Backport transposes optimization to v0.3.0 (#3994)

* Optimizer: optimize transposes in variety of circumstances (#3509)

* Optimizer: Optimize transposes in variety of circumstances

- No-op transposes
- Consecutive transposes (fuse them)
- Transposes into Gemm (fuse them into transA/transB parameter)

* touch up out of date comment

* Backporting optimizer changes

README.md

@@ -202,9 +202,9 @@ MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py install
 Dockerfile is supplied to build images with cuda support and cudnn v6. Build as usual
 ```
 docker build -t pytorch .
-
+```
 Dockerfile to build with cuda 9 and cudnn v7 (with Volta support) is in tools/docker, the build command is
-
+```
 docker build -t pytorch_cuda9 -f tools/docker/Dockerfile9 .
 ```
 Alternatively, if you want to use a runtime image, you can use the pre-built one from Docker Hub and run with nvidia-docker:

docs/source/autograd.rst

@@ -56,6 +56,12 @@ gradients are correct.
 Profiler
 --------
 
+Autograd includes a profiler that lets you inspect the cost of different
+operators inside your model - both on the CPU and GPU. There are two modes
+implemented at the moment - CPU-only using :class:`~torch.autograd.profiler.profile`.
+and nvprof based (registers both CPU and GPU activity) using
+:class:`~torch.autograd.profiler.emit_nvtx`.
+
 .. autoclass:: torch.autograd.profiler.profile
     :members:
 

docs/source/cuda.rst

@@ -37,6 +37,10 @@ Streams and events
 .. autoclass:: Event
    :members:
 
+Memory management
+-----------------
+.. autofunction:: empty_cache
+
 NVIDIA Tools Extension (NVTX)
 -----------------------------
 

docs/source/distributions.rst

@@ -19,10 +19,10 @@ Probability distributions - torch.distributions
 .. autoclass:: Bernoulli
     :members:
 
-:hidden:`Multinomial`
+:hidden:`Categorical`
 ~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: Multinomial
+.. autoclass:: Categorical
     :members:
 
 :hidden:`Normal`

docs/source/notes/cuda.rst

@@ -3,18 +3,19 @@
 CUDA semantics
 ==============
 
-:mod:`torch.cuda` keeps track of currently selected GPU, and all CUDA tensors
-you allocate will be created on it. The selected device can be changed with a
+:mod:`torch.cuda` is used to set up and run CUDA operations. It keeps track of
+the currently selected GPU, and all CUDA tensors you allocate will by default be
+created on that device. The selected device can be changed with a
 :any:`torch.cuda.device` context manager.
 
-However, once a tensor is allocated, you can do operations on it irrespectively
-of your selected device, and the results will be always placed in on the same
+However, once a tensor is allocated, you can do operations on it irrespective
+of the selected device, and the results will be always placed in on the same
 device as the tensor.
 
 Cross-GPU operations are not allowed by default, with the only exception of
-:meth:`~torch.Tensor.copy_`. Unless you enable peer-to-peer memory accesses,
-any attempts to launch ops on tensors spread across different devices will
-raise an error.
+:meth:`~torch.Tensor.copy_`. Unless you enable peer-to-peer memory access, any
+attempts to launch ops on tensors spread across different devices will raise an
+error.
 
 Below you can find a small example showcasing this::
 
@@ -41,6 +42,16 @@ Below you can find a small example showcasing this::
         d = torch.randn(2).cuda(2)
         # d.get_device() == 2
 
+Memory management
+-----------------
+
+PyTorch use a caching memory allocator to speed up memory allocations. This
+allows fast memory deallocation without device synchronizations. However, the
+unused memory managed by the allocator will still show as if used in
+`nvidia-smi`. Calling :meth:`~torch.cuda.empty_cache` can release all unused
+cached memory from PyTorch so that those can be used by other GPU applications.
+
+
 Best practices
 --------------
 
@@ -49,13 +60,13 @@ Device-agnostic code
 
 Due to the structure of PyTorch, you may need to explicitly write
 device-agnostic (CPU or GPU) code; an example may be creating a new tensor as
-the initial hidden state of a recurrent neural network. 
+the initial hidden state of a recurrent neural network.
 
 The first step is to determine whether the GPU should be used or not. A common
-pattern is to use Python's `argparse` module to read in user arguments, and
+pattern is to use Python's ``argparse`` module to read in user arguments, and
 have a flag that can be used to disable CUDA, in combination with
-`torch.cuda.is_available()`. In the following, `args.cuda` results in a flag
-that can be used to cast tensors and modules to CUDA if desired::
+:meth:`~torch.cuda.is_available`. In the following, ``args.cuda`` results in a
+flag that can be used to cast tensors and modules to CUDA if desired::
 
     import argparse
     import torch
@@ -66,7 +77,7 @@ that can be used to cast tensors and modules to CUDA if desired::
     args = parser.parse_args()
     args.cuda = not args.disable_cuda and torch.cuda.is_available()
 
-If modules or tensors need to be sent to the GPU, `args.cuda` can be used as
+If modules or tensors need to be sent to the GPU, ``args.cuda`` can be used as
 follows::
 
     x = torch.Tensor(8, 42)
@@ -84,9 +95,9 @@ dataloader would be as follows::
         x = Variable(x.type(dtype))
 
 When working with multiple GPUs on a system, you can use the
-`CUDA_VISIBLE_DEVICES` environment flag to manage which GPUs are available to
-PyTorch. To manually control which GPU a tensor is created on, the best practice
-is to use the `torch.cuda.device()` context manager::
+``CUDA_VISIBLE_DEVICES`` environment flag to manage which GPUs are available to
+PyTorch. As mentioned above, to manually control which GPU a tensor is created
+on, the best practice is to use a :any:`torch.cuda.device` context manager::
 
     print("Outside device is 0")  # On device 0 (default in most scenarios)
     with torch.cuda.device(1):
@@ -94,9 +105,10 @@ is to use the `torch.cuda.device()` context manager::
     print("Outside device is still 0")  # On device 0
 
 If you have a tensor and would like to create a new tensor of the same type on
-the same device, then you can use the `.new()` function, which acts the same as
-a normal tensor constructor. Whilst the previously mentioned methods depend on
-the current GPU context, `new()` preserves the device of the original tensor.
+the same device, then you can use the :meth:`~torch.Tensor.new` method, which
+acts the same as a normal tensor constructor. Whilst the previously mentioned
+methods depend on the current GPU context, :meth:`~torch.Tensor.new` preserves
+the device of the original tensor.
 
 This is the recommended practice when creating modules in which new
 tensors/variables need to be created internally during the forward pass::
@@ -110,8 +122,9 @@ tensors/variables need to be created internally during the forward pass::
     y_cpu_long = x_cpu_long.new([[1, 2, 3]])
 
 If you want to create a tensor of the same type and size of another tensor, and
-fill it with either ones or zeros, `torch.ones_like()` or `torch.zeros_like()`
-are provided as more convenient functions (which also preserve device)::
+fill it with either ones or zeros, :meth:`~torch.ones_like` or
+:meth:`~torch.zeros_like` are provided as convenient helper functions (which
+also preserve device)::
 
     x_cpu = torch.FloatTensor(1)
     x_gpu = torch.cuda.FloatTensor(1)
@@ -145,9 +158,9 @@ pinned memory by passing ``pin_memory=True`` to its constructor.
 Use nn.DataParallel instead of multiprocessing
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-Most use cases involving batched input and multiple GPUs should default to using
-:class:`~torch.nn.DataParallel` to utilize more than one GPU. Even with the GIL,
-a single python process can saturate multiple GPUs.
+Most use cases involving batched inputs and multiple GPUs should default to
+using :class:`~torch.nn.DataParallel` to utilize more than one GPU. Even with
+the GIL, a single Python process can saturate multiple GPUs.
 
 As of version 0.1.9, large numbers of GPUs (8+) might not be fully utilized.
 However, this is a known issue that is under active development. As always,

docs/source/onnx.rst

@@ -53,7 +53,7 @@ exporter to print out a human-readable representation of the network::
 You can also verify the protobuf using the `onnx <https://github.com/onnx/onnx/>`_ library.
 You can install ``onnx`` with conda::
 
-    conda install -c ezyang onnx
+    conda install -c conda-forge onnx
 
 Then, you can run::
 
@@ -75,10 +75,8 @@ To run the exported script with `caffe2 <https://caffe2.ai/>`_, you will need th
 
 2. You'll need `onnx-caffe2 <https://github.com/onnx/onnx-caffe2>`_, a
    pure-Python library which provides a Caffe2 backend for ONNX.  You can install ``onnx-caffe2``
-   with conda or pip::
+   with pip::
 
-      conda install -c ezyang onnx-caffe2
-      # OR
       pip install onnx-caffe2
 
 Once these are installed, you can use the backend for Caffe2::
@@ -122,34 +120,48 @@ Limitations
 Supported operators
 -------------------
 
-In this tech preview, only the following operators are supported:
+The following operators are supported:
 
-* Add (inplace is discarded)
-* Sub (inplace is discarded)
-* Mul (inplace is discarded)
-* Negate (inplace is discarded)
-* Addmm (inplace is discarded, alpha and beta must be 1)
-* Tanh (inplace is discarded)
-* Sigmoid (inplace is discarded)
-* Transpose
-* View
-* Permute
-* Concat
-* Squeeze (inplace is discarded)
+* add (nonzero alpha not supported)
+* sub (nonzero alpha not supported)
+* mul
+* div
+* cat
+* mm
+* addmm
+* neg
+* tanh
+* sigmoid
+* mean
+* t
+* expand (only when used before a broadcasting ONNX operator; e.g., add)
+* transpose
+* view
+* split
+* squeeze
+* prelu (single weight shared among input channels not supported)
+* threshold (non-zero threshold/non-zero value not supported)
+* leaky_relu
+* glu
+* softmax
+* avg_pool2d (ceil_mode not supported)
+* log_softmax
+* unfold (experimental support with ATen-Caffe2 integration)
+* elu
+* Conv
 * BatchNorm
-* Convolution
-* Embedding (only optional argument that is supported is ``padding_idx``)
-* Slice (only integer indexing is supported)
-* Dropout (inplace is discarded)
-* Relu (inplace is discarded)
-* PReLU (inplace is discarded, sharing a single weight among all channels is not supported)
-* LeakyRelu (inplace is discarded)
-* MaxPool1d (ceil_mode must be False)
-* MaxPool2d (ceil_mode must be False)
-* AvgPool2d (ceil_mode must be False)
+* MaxPool1d (ceil_mode not supported)
+* MaxPool2d (ceil_mode not supported)
+* MaxPool3d (ceil_mode not supported)
+* Embedding (no optional arguments supported)
+* RNN
+* ConstantPadNd
+* Dropout
+* FeatureDropout (training mode not supported)
+* Index (constant integer and tuple indices supported)
+* Negate
 
-We plan on expanding support to more operators; RNNs are high on our priority
-list.  The operator set above is sufficient to export the following models:
+The operator set above is sufficient to export the following models:
 
 * AlexNet
 * DCGAN

docs/source/optim.rst

@@ -18,11 +18,11 @@ you can specify optimizer-specific options such as the learning rate, weight dec
 
 .. note::
 
-    If you need to move a model to GPU via `.cuda()`, please do so before 
+    If you need to move a model to GPU via `.cuda()`, please do so before
     constructing optimizers for it. Parameters of a model after `.cuda()` will
-    be different objects with those before the call. 
+    be different objects with those before the call.
 
-    In general, you should make sure that optimized parameters live in  
+    In general, you should make sure that optimized parameters live in
     consistent locations when optimizers are constructed and used.
 
 Example::
@@ -111,6 +111,8 @@ Algorithms
     :members:
 .. autoclass:: Adam
     :members:
+.. autoclass:: SparseAdam
+    :members:
 .. autoclass:: Adamax
     :members:
 .. autoclass:: ASGD

setup.py

@@ -542,7 +542,7 @@ if os.getenv('PYTORCH_BINARY_BUILD') and platform.system() == 'Linux':
     STDCPP_LIB = STDCPP_LIB[:-1]
     if type(STDCPP_LIB) != str:  # python 3
         STDCPP_LIB = STDCPP_LIB.decode(sys.stdout.encoding)
-    main_link_args += [STDCPP_LIB]
+    extra_link_args += [STDCPP_LIB]
     version_script = os.path.abspath("tools/pytorch.version")
     extra_link_args += ['-Wl,--version-script=' + version_script]
 
@@ -593,9 +593,11 @@ extensions.append(THNN)
 if WITH_CUDA:
     thnvrtc_link_flags = extra_link_args + [make_relative_rpath('lib')]
     if platform.system() == 'Linux':
-        thnvrtc_link_flags = ['-Wl,--no-as-needed'] + thnvrtc_link_flags
+        thnvrtc_link_flags = thnvrtc_link_flags + ['-Wl,--no-as-needed']
+    # these have to be specified as -lcuda in link_flags because they
+    # have to come right after the `no-as-needed` option
+    thnvrtc_link_flags += ['-lcuda', '-lnvrtc']
     THNVRTC = Extension("torch._nvrtc",
-                        libraries=['nvrtc', 'cuda'],
                         sources=['torch/csrc/nvrtc.cpp'],
                         language='c++',
                         include_dirs=include_dirs,
@@ -618,11 +620,13 @@ if WITH_CUDA:
                        )
     extensions.append(THCUNN)
 
-version = '0.2.0'
+version = '0.3.0b0'
 if os.getenv('PYTORCH_BUILD_VERSION'):
     assert os.getenv('PYTORCH_BUILD_NUMBER') is not None
-    version = os.getenv('PYTORCH_BUILD_VERSION') \
-        + '_' + os.getenv('PYTORCH_BUILD_NUMBER')
+    build_number = int(os.getenv('PYTORCH_BUILD_NUMBER'))
+    version = os.getenv('PYTORCH_BUILD_VERSION')
+    if build_number > 1:
+        version += '.post' + str(build_number)
 else:
     try:
         sha = subprocess.check_output(['git', 'rev-parse', 'HEAD'], cwd=cwd).decode('ascii').strip()

test/common.py

@@ -170,6 +170,9 @@ class TestCase(unittest.TestCase):
         return x, y
 
     def assertEqual(self, x, y, prec=None, message=''):
+        if isinstance(prec, str) and message == '':
+            message = prec
+            prec = None
         if prec is None:
             prec = self.precision
 

test/common_nn.py

@@ -246,6 +246,17 @@ module_tests = [
 ]
 
 
+def kldivloss_reference(input, target, size_average=True, reduce=True):
+    safe_target = target * (target > 0).type_as(target)
+    safe_target_log = (safe_target + (target <= 0).type_as(target)).log()
+    result = safe_target * (safe_target_log - input)
+    if reduce and size_average:
+        return result.mean()
+    elif reduce:
+        return result.sum()
+    return result
+
+
 def nllloss2d_reference(input, target, weight=None, ignore_index=-100,
                         size_average=True, reduce=True):
     N, C, H, W = input.size()
@@ -309,6 +320,7 @@ def smoothl1loss_reference(input, target, size_average=True, reduce=True):
 
 
 loss_reference_fns = {
+    'KLDivLoss': kldivloss_reference,
     'NLLLoss': nllloss_reference,
     'NLLLoss2d': nllloss2d_reference,
     'SmoothL1Loss': smoothl1loss_reference,
@@ -370,6 +382,8 @@ criterion_tests = [
         module_name='KLDivLoss',
         input_fn=lambda: torch.rand(10, 10).log(),
         target_fn=lambda: torch.rand(10, 10),
+        reference_fn=lambda i, t, m:
+            kldivloss_reference(i, t, get_size_average(m), reduce=True),
         check_no_size_average=True,
     ),
     dict(

test/test_autograd.py

@@ -995,6 +995,16 @@ class TestAutograd(TestCase):
         self._test_setitem_tensor((5, 5), Variable(mask))
         self._test_setitem_tensor((5,), Variable(mask[0]))
 
+    def test_select_sum(self):
+        # both select and sum return Scalars in ATen; ensure they work together.
+        x = Variable(torch.randn(10), requires_grad=True)
+
+        def func(x):
+            return x.select(0, 1).sum()
+
+        gradcheck(func, [x])
+        gradgradcheck(func, [x])
+
     def test_stack(self):
         x = Variable(torch.randn(10, 10), requires_grad=True)
         y = Variable(torch.randn(10, 10), requires_grad=True)
@@ -1006,6 +1016,43 @@ class TestAutograd(TestCase):
         self.assertEqual(y.grad.data, grad[1])
         self.assertEqual(z.grad.data, grad[2])
 
+    def test_put(self):
+        root = Variable(torch.randn(4, 5), requires_grad=True)
+        values = Variable(torch.randn(6), requires_grad=True)
+        idx = Variable(torch.LongTensor([1, 2, 3, -1, -2, -3]))
+
+        def func(root, values):
+            x = root.clone()
+            x.put_(idx, values)
+            return x
+
+        gradcheck(func, [root, values])
+        gradgradcheck(func, [root, values])
+
+    def test_put_accumulate(self):
+        root = Variable(torch.randn(4, 5), requires_grad=True)
+        values = Variable(torch.randn(6), requires_grad=True)
+        idx = Variable(torch.LongTensor([1, 2, 3, 1, 2, 3]))
+
+        def func(root, values):
+            x = root.clone()
+            x.put_(idx, values, accumulate=True)
+            return x
+
+        gradcheck(func, [root, values])
+        gradgradcheck(func, [root, values])
+
+    def test_fill(self):
+        root = Variable(torch.randn(4, 5), requires_grad=True)
+
+        def func(root):
+            x = root.clone()
+            x.fill_(2)
+            return x
+
+        gradcheck(func, [root])
+        gradgradcheck(func, [root])
+
     def test_unused_output(self):
         x = Variable(torch.randn(10, 10), requires_grad=True)
         outputs = x.chunk(5)
@@ -1461,13 +1508,14 @@ class TestAutograd(TestCase):
     def test_norm_subgradient(self):
         def run_test(input_size, norm_deg):
             input = Variable(torch.zeros(*input_size), requires_grad=True)
-            out = input.norm(norm_deg)
-            out.backward()
+            input.norm(norm_deg).backward()
             self.assertEqual(input.grad.data.abs().sum(), 0)
 
         run_test((10,), 2)
         run_test((10, 10), 2)
         run_test((10,), 3)
+        run_test((10,), 1)
+        run_test((10,), 1.5)
 
     def test_profiler(self):
         x = Variable(torch.randn(10, 10))
@@ -1764,8 +1812,14 @@ method_tests = [
     ('addcdiv', (S, S), (0.5, (S, 1), (1, S)), 'scale_broadcast_rhs'),
     ('addcdiv', (1,), (0.5, (S, S, 1), (1, S)), 'scale_broadcast_all'),
     ('zero_', (S, S, S), ()),
-    ('norm', (S, S, S), (2,)),
-    ('norm', (S, S, S), (3,), '3'),
+    ('norm', (S, S), (2,)),
+    ('norm', (S, S), (0,), '0'),
+    ('norm', (S, S), (0.5,), '0_5'),
+    ('norm', (S, S), (1,), '1'),
+    ('norm', (S, S), (3,), '3'),
+    ('norm', (S, S), (-1,), 'neg_1'),
+    ('norm', (S, S), (-0.5,), 'neg_0_5'),
+    ('norm', (S, S), (-1.5,), 'neg_1_5'),
     ('norm', torch.rand(S, S, S) + 5e-2, (1.5,), '1_5'),
     ('norm', (S, S, S), (2, 1), '2_dim', [1]),
     ('norm', (S, S, S), (3, 1), '3_dim', [1]),
@@ -1842,6 +1896,7 @@ method_tests = [
     ('squeeze', (S, 1, S, 1), ()),
     ('squeeze', (S, 1, S, 1), (1,), '1_dim', [0]),
     ('squeeze', (S, 1, S, 1), (2,), 'not_1_dim', [0]),
+    ('squeeze', (1,), (0,), '1d_dim0', [0]),
     ('unsqueeze', (S, S, S), (0,), 'first', [0]),
     ('unsqueeze', (S, S, S), (1,), 'middle', [0]),
     ('unsqueeze', (S, S, S), (3,), 'last', [0]),
@@ -1875,6 +1930,7 @@ method_tests = [
     ('topk', (S, M, S), (3, 1), 'dim'),
     ('topk', (S, M, S), (3, 1, True), 'dim_desc'),
     ('topk', (S, M, S), (3, 1, True, True), 'dim_desc_sort'),
+    ('take', (S, S, S), (Variable(torch.LongTensor([[-3, 2], [20, 2]])),)),
     ('__getitem__', torch.randn(S, S, S), (dont_convert([1, 2]),)),
     ('__getitem__', torch.randn(S, S, S), (slice(0, 3),), 'slice'),
     ('__getitem__', torch.randn(S, S, S), (dont_convert([slice(0, 3), 1]),), 'slice_index'),

test/test_cuda.py

@@ -1,5 +1,6 @@
 import math
 import tempfile
+import re
 import unittest
 from itertools import repeat
 
@@ -16,6 +17,11 @@ if not torch.cuda.is_available():
     TestCase = object  # noqa: F811
     HAS_CUDA = False
 
+HAS_MAGMA = HAS_CUDA
+if HAS_CUDA:
+    torch.ones(1).cuda()  # has_magma shows up after cuda is initialized
+    HAS_MAGMA = torch.cuda.has_magma
+
 
 def is_floating(t):
     return type(t) in [torch.FloatTensor, torch.DoubleTensor,
@@ -968,6 +974,69 @@ class TestCuda(TestCase):
     def test_tensor_scatterFill(self):
         TestTorch._test_scatter_base(self, lambda t: t.cuda(), 'scatter_', True, test_bounds=False)
 
+    def test_var(self):
+        cpu_tensor = torch.randn(2, 3, 3)
+        gpu_tensor = cpu_tensor.cuda()
+        self.assertEqual(gpu_tensor.var(), cpu_tensor.var())
+        self.assertEqual(gpu_tensor.var(1), cpu_tensor.var(1))
+        self.assertEqual(gpu_tensor.var(2), cpu_tensor.var(2))
+        self.assertEqual(gpu_tensor.std(), cpu_tensor.std())
+        self.assertEqual(gpu_tensor.std(1), cpu_tensor.std(1))
+        self.assertEqual(gpu_tensor.var(2), cpu_tensor.var(2))
+
+        cpu_tensor = torch.randn(100)
+        gpu_tensor = cpu_tensor.cuda()
+        self.assertEqual(gpu_tensor.var(), cpu_tensor.var())
+
+    def test_var_unbiased(self):
+        tensor = torch.randn(100).cuda()
+        self.assertEqual(tensor.var(0), tensor.var(0, unbiased=True))
+        self.assertEqual(tensor.var(), tensor.var(unbiased=True))
+        self.assertEqual(tensor.var(unbiased=False), tensor.var(0, unbiased=False)[0])
+
+        tensor = torch.FloatTensor([1.0, 2.0]).cuda()
+        self.assertEqual(tensor.var(unbiased=True), 0.5)
+        self.assertEqual(tensor.var(unbiased=False), 0.25)
+
+        tensor = torch.randn(100).cuda()
+        self.assertEqual(tensor.std(0), tensor.std(0, unbiased=True))
+        self.assertEqual(tensor.std(), tensor.std(unbiased=True))
+        self.assertEqual(tensor.std(unbiased=False), tensor.std(0, unbiased=False)[0])
+
+    def test_var_large_input(self):
+        # Large, not-nice input
+        tensor_cpu = torch.randn(2 * 32 * 1024 + 1, 2, 67)
+        tensor_cuda = tensor_cpu.cuda()
+
+        self.assertEqual(tensor_cpu.var(2), tensor_cuda.var(2).cpu())
+
+    def test_var_stability(self):
+        tensor = torch.FloatTensor([2281.5, 2281.25]).cuda()
+
+        # Stability for inner dim
+        self.assertEqual(tensor.var(0)[0], 0.03125)
+
+        # General stability
+        self.assertEqual(tensor.var(), 0.03125)
+
+        # Stability for outer dimensions
+        tensor = tensor.unsqueeze(1)
+        self.assertEqual(tensor.var(0)[0], 0.03125)
+
+    @unittest.skipIf(not HAS_MAGMA, "no MAGMA library detected")
+    def test_symeig(self):
+        # Small case
+        tensor = torch.randn(3, 3).cuda()
+        tensor = torch.mm(tensor, tensor.t())
+        eigval, eigvec = torch.symeig(tensor, eigenvectors=True)
+        self.assertEqual(tensor, torch.mm(torch.mm(eigvec, eigval.diag()), eigvec.t()))
+
+        # Large case
+        tensor = torch.randn(257, 257).cuda()
+        tensor = torch.mm(tensor, tensor.t())
+        eigval, eigvec = torch.symeig(tensor, eigenvectors=True)
+        self.assertEqual(tensor, torch.mm(torch.mm(eigvec, eigval.diag()), eigvec.t()))
+
     def test_arange(self):
         for t in ['IntTensor', 'LongTensor', 'FloatTensor', 'DoubleTensor']:
             a = torch.cuda.__dict__[t]()

test/test_dataloader.py

@@ -4,6 +4,7 @@ import torch
 import traceback
 import unittest
 from torch.utils.data import Dataset, TensorDataset, DataLoader, ConcatDataset
+from torch.utils.data.dataloader import default_collate
 from common import TestCase, run_tests, TEST_NUMPY
 from common_nn import TEST_CUDA
 
@@ -276,6 +277,23 @@ class TestDataLoader(TestCase):
             batch = next(iter(loader))
             self.assertIsInstance(batch, tt)
 
+    @unittest.skipIf(not TEST_NUMPY, "numpy unavailable")
+    def test_default_colate_bad_numpy_types(self):
+        import numpy as np
+
+        # Should be a no-op
+        arr = np.array(['a', 'b', 'c'])
+        default_collate(arr)
+
+        arr = np.array([[['a', 'b', 'c']]])
+        self.assertRaises(TypeError, lambda: default_collate(arr))
+
+        arr = np.array([object(), object(), object()])
+        self.assertRaises(TypeError, lambda: default_collate(arr))
+
+        arr = np.array([[[object(), object(), object()]]])
+        self.assertRaises(TypeError, lambda: default_collate(arr))
+
 
 class StringDataset(Dataset):
     def __init__(self):

test/test_distributions.py

@@ -2,7 +2,7 @@ from common import TestCase, run_tests
 import math
 import torch
 from torch.autograd import Variable, gradcheck
-from torch.distributions import Bernoulli, Multinomial, Normal
+from torch.distributions import Bernoulli, Categorical, Normal
 
 
 class TestDistributions(TestCase):
@@ -47,22 +47,22 @@ class TestDistributions(TestCase):
     def test_multinomial_1d(self):
         p = Variable(torch.Tensor([0.1, 0.2, 0.3]), requires_grad=True)
         # TODO: this should return a 0-dim tensor once we have Scalar support
-        self.assertEqual(Multinomial(p).sample().size(), (1,))
-        self.assertEqual(Multinomial(p).sample_n(1).size(), (1, 1))
-        self._gradcheck_log_prob(Multinomial, (p,))
+        self.assertEqual(Categorical(p).sample().size(), (1,))
+        self.assertEqual(Categorical(p).sample_n(1).size(), (1, 1))
+        self._gradcheck_log_prob(Categorical, (p,))
 
     def test_multinomial_2d(self):
         probabilities = [[0.1, 0.2, 0.3], [0.5, 0.3, 0.2]]
         p = Variable(torch.Tensor(probabilities), requires_grad=True)
-        self.assertEqual(Multinomial(p).sample().size(), (2,))
-        self.assertEqual(Multinomial(p).sample_n(6).size(), (6, 2))
-        self._gradcheck_log_prob(Multinomial, (p,))
+        self.assertEqual(Categorical(p).sample().size(), (2,))
+        self.assertEqual(Categorical(p).sample_n(6).size(), (6, 2))
+        self._gradcheck_log_prob(Categorical, (p,))
 
         def ref_log_prob(idx, val, log_prob):
             sample_prob = p.data[idx][val] / p.data[idx].sum()
             self.assertEqual(log_prob, math.log(sample_prob))
 
-        self._check_log_prob(Multinomial(p), ref_log_prob)
+        self._check_log_prob(Categorical(p), ref_log_prob)
 
     def test_normal(self):
         mean = Variable(torch.randn(5, 5), requires_grad=True)

test/test_jit.py

@@ -15,6 +15,15 @@ try:
 except ImportError:
     HAS_TORCHVISION = False
 
+RUN_CUDA = torch.cuda.is_available()
+if torch.cuda.is_available():
+    CUDA_VERSION = torch._C._cuda_getCompiledVersion()
+    for d in range(torch.cuda.device_count()):
+        major = torch.cuda.get_device_capability(d)[0]
+        if (CUDA_VERSION < 8000 and major >= 6) or (CUDA_VERSION < 9000 and major >= 7):
+            RUN_CUDA = False
+
+
 skipIfNoTorchVision = unittest.skipIf(not HAS_TORCHVISION, "no torchvision")
 
 
@@ -52,7 +61,7 @@ class TestJit(TestCase):
         torch._C._jit_pass_lint(trace)
         self.assertExpected(str(trace))
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_lstm_fusion(self):
         input = Variable(torch.randn(3, 10).cuda())
         hx = Variable(torch.randn(3, 20).cuda())
@@ -65,7 +74,7 @@ class TestJit(TestCase):
         torch._C._jit_pass_lint(trace)
         self.assertExpected(str(trace))
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_run_lstm_fusion(self):
         input = Variable(torch.randn(3, 10).cuda())
         hx = Variable(torch.randn(3, 20).cuda())
@@ -78,7 +87,7 @@ class TestJit(TestCase):
         z2 = CompiledLSTMCell(input, (hx, cx), *module.parameters(), _assert_compiled=True)
         self.assertEqual(z, z2)
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_run_lstm_fusion_concat(self):
         input = Variable(torch.randn(3, 10).cuda())
         hx = Variable(torch.randn(3, 20).cuda())
@@ -91,7 +100,7 @@ class TestJit(TestCase):
         z2 = CompiledLSTMCell(input, (hx, cx), *module.parameters(), _assert_compiled=True)
         self.assertEqual(z, z2)
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_concat_fusion(self):
         hx = Variable(torch.randn(3, 20).cuda())
         cx = Variable(torch.randn(3, 20).cuda())
@@ -105,7 +114,7 @@ class TestJit(TestCase):
         torch._C._jit_pass_lint(trace)
         self.assertExpected(str(trace))
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_fusion_distribute(self):
         def f(x, y):
             z1, z2 = (x + y).chunk(2, dim=1)
@@ -146,7 +155,7 @@ class TestJit(TestCase):
         self.assertEqual(z, torch.sigmoid(torch.tanh(x * (x + y))))
         self.assertEqual(z, z2)
 
-    @unittest.skipIf(not torch.cuda.is_available(), "fuser requires CUDA")
+    @unittest.skipIf(not RUN_CUDA, "fuser requires CUDA")
     def test_compile_addc(self):
         x = Variable(torch.Tensor([0.4]), requires_grad=True).cuda()
         y = Variable(torch.Tensor([0.7]), requires_grad=True).cuda()
@@ -613,7 +622,7 @@ class TestJit(TestCase):
         assert(torch.equal(torch.ones([2, 2]), t_node.t("a")))
         self.assertExpected(str(g2))
 
-    @unittest.skipIf(not torch.cuda.is_available(), "cpp tests require CUDA")
+    @unittest.skipIf(not RUN_CUDA, "cpp tests require CUDA")
     def test_cpp(self):
         torch._C._jit_run_cpp_tests()
 

test/test_nn.py

@@ -2249,6 +2249,38 @@ class TestNN(NNTestCase):
             weight_data[:] = 4
             self.assertEqual(weight_data, all_vars[4].data)
 
+    @unittest.skipIf(not TEST_CUDNN, 'CUDNN not available')
+    def test_cudnn_weight_tying(self):
+        rnns = [
+            nn.LSTM(10, 20, batch_first=True, bidirectional=True),
+            nn.GRU(10, 20, batch_first=True, bidirectional=True),
+            nn.RNN(10, 20, batch_first=True, bidirectional=True)
+        ]
+        for rnn in rnns:
+            rnn.bias_ih_l0_reverse = rnn.bias_ih_l0
+            rnn.cuda()
+            input = Variable(torch.randn(5, 4, 10).cuda(), requires_grad=True)
+            hx = Variable(torch.randn(2, 5, 20).cuda(), requires_grad=True)
+            all_vars = [input, hx] + list(rnn.parameters())
+            opt = torch.optim.SGD(rnn.parameters(), lr=0.1)
+            opt.zero_grad()
+            if isinstance(rnn, nn.LSTM):
+                cx = Variable(torch.randn(2, 5, 20).cuda(), requires_grad=True)
+                all_vars[2:2] = [cx]
+                hx = (hx, cx)
+
+            with warnings.catch_warnings(record=True) as w:
+                output = rnn(input, hx)
+            output[0].sum().backward()
+
+            opt.step()
+            with warnings.catch_warnings(record=True) as w:
+                output_cuda = rnn(input, hx)
+            rnn.cpu()
+            hx = (hx[0].cpu(), hx[1].cpu()) if isinstance(rnn, nn.LSTM) else hx.cpu()
+            output_cpu = rnn(input.cpu(), hx)
+            self.assertEqual(output_cuda, output_cpu)
+
     @unittest.skipIf(not TEST_CUDA, 'CUDA not available')
     def test_cuda_rnn_fused(self):
         def copy_rnn(rnn1, rnn2):
@@ -2759,6 +2791,26 @@ class TestNN(NNTestCase):
 
         self.assertEqual(out1, out2)
 
+    def test_elu_inplace_gradgrad(self):
+        v = Variable(torch.randn(8), requires_grad=True)
+
+        def func(root):
+            x = root.clone()
+            return F.elu(x, inplace=True)
+
+        gradcheck(func, [v])
+        gradgradcheck(func, [v])
+
+    def test_hardtanh_inplace_gradgrad(self):
+        v = Variable(torch.randn(8), requires_grad=True)
+
+        def func(root):
+            x = root.clone()
+            return F.hardtanh(x, inplace=True)
+
+        gradcheck(func, [v])
+        gradgradcheck(func, [v])
+
     def test_batchnorm_raises_error_if_running_mean_is_not_same_size_as_input(self):
         input = Variable(torch.rand(2, 10))
         running_var = torch.rand(10)
@@ -2845,38 +2897,17 @@ class TestNN(NNTestCase):
         self.assertTrue(gradcheck(lambda x, y: F.cosine_similarity(x, y, dim=-1), (input1, input2)))
 
     def test_grid_sample(self):
-        # test known input on CPU
-        input = Variable(torch.arange(1, 11).view(1, 1, 2, 5))
-        grid = Variable(torch.Tensor(
-            [[-1, -0.5, 0, 0.2, 1],
-             [-1, -0.333, 0, 0.5, 1],
-             [-1, -0.5, 0, 0.3333, 1],
-             [-1, -0.2, 0, 0.2, 1]]).view(1, 2, 5, 2))
-        output = F.grid_sample(input, grid)
-        groundtruth = torch.Tensor(
-            [[2.2500, 6.0000000000, 5.0000, 4.8340, 9.0000],
-             [2.2500, 6.333250045, 5.0000, 5.1000, 8.4000]]).view(1, 1, 2, 5)
-        self.assertEqual(output.data, groundtruth)
+        def test_cpu_against_cuda(N, C, H, W, padding_mode):
+            def test_shape(N, C, IH, IW, H, W, padding_mode):
 
-        # do gradcheck
-        N = random.randint(1, 8)
-        C = random.randint(1, 8)
-        H = random.randint(1, 8)
-        W = random.randint(1, 8)
-        input = Variable(torch.randn(N, C, H, W), requires_grad=True)
-        grid = Variable(torch.randn(N, H, W, 2), requires_grad=True)
-        self.assertTrue(gradcheck(lambda inp, grid: F.grid_sample(inp, grid), (input, grid)))
-
-        def test_cpu_against_cuda(N, C, H, W):
-            def test_shape(N, C, IH, IW, H, W):
                 input_cpu = Variable(torch.randn(C, N, IH, IW).transpose(0, 1), requires_grad=True)
                 grid_cpu = Variable(torch.randn(H, N, W, 2).transpose(0, 1), requires_grad=True)
-                out_cpu = F.grid_sample(input_cpu, grid_cpu)
+                out_cpu = F.grid_sample(input_cpu, grid_cpu, padding_mode=padding_mode)
                 self.assertTrue(out_cpu.size() == torch.Size([N, C, H, W]))
 
                 input_cuda = Variable(input_cpu.data.transpose(0, 1).cuda().transpose(0, 1), requires_grad=True)
                 grid_cuda = Variable(grid_cpu.data.transpose(0, 1).cuda().transpose(0, 1), requires_grad=True)
-                out_cuda = F.grid_sample(input_cuda, grid_cuda)
+                out_cuda = F.grid_sample(input_cuda, grid_cuda, padding_mode=padding_mode)
                 self.assertEqual(out_cpu, out_cuda)
 
                 gradients = out_cpu.data.new(out_cpu.size()).normal_()
@@ -2889,15 +2920,15 @@ class TestNN(NNTestCase):
                 base_input = torch.randn(C, IH, IW)
                 input_cpu = Variable(base_input.expand(input_cuda.size()), requires_grad=True)
                 grid_cpu = Variable(torch.randn(N, H, W, 2), requires_grad=True)
-                out_cpu = F.grid_sample(input_cpu, grid_cpu)
+                out_cpu = F.grid_sample(input_cpu, grid_cpu, padding_mode=padding_mode)
 
                 input_cuda = Variable(base_input.cuda().expand(input_cuda.size()), requires_grad=True)
                 grid_cuda = Variable(grid_cpu.data.cuda(), requires_grad=True)
-                out_cuda = F.grid_sample(input_cuda, grid_cuda)
+                out_cuda = F.grid_sample(input_cuda, grid_cuda, padding_mode=padding_mode)
                 self.assertEqual(out_cpu, out_cuda)
 
             # test same size output
-            test_shape(N, C, H, W, H, W)
+            test_shape(N, C, H, W, H, W, padding_mode)
 
             # test larger output
             N = random.randint(1, 8)
@@ -2906,7 +2937,7 @@ class TestNN(NNTestCase):
             IW = random.randint(1, 8)
             H = random.randint(IH + 1, 12)
             W = random.randint(IH + 1, 12)
-            test_shape(N, C, IH, IW, H, W)
+            test_shape(N, C, IH, IW, H, W, padding_mode)
 
             # test smaller output
             N = random.randint(1, 8)
@@ -2915,21 +2946,44 @@ class TestNN(NNTestCase):
             IW = random.randint(1, 8)
             H = random.randint(1, IH)
             W = random.randint(1, IW)
-            test_shape(N, C, IH, IW, H, W)
+            test_shape(N, C, IH, IW, H, W, padding_mode)
 
-        # test CUDNN against CPU
-        if TEST_CUDNN:
-            test_cpu_against_cuda(N, C, H, W)
+        # test known input on CPU
+        for padding_mode in ['zeros', 'border']:
 
-        # test CUDA (without CUDNN) against CPU
-        if TEST_CUDA:
+            input = Variable(torch.arange(1, 11).view(1, 1, 2, 5))
+            grid = Variable(torch.Tensor(
+                [[-0.9, -1.4, 0, 0.2, 1],
+                 [-1, -0.333, 0, 0.5, 1],
+                 [-1, -0.5, 0, 0.3333, 1],
+                 [-1, -0.2, 0, 1.1, 0.5]]).view(1, 2, 5, 2))
+            output = F.grid_sample(input, grid, padding_mode=padding_mode)
 
-            # GridSampler will automatically use CUDNN if it is available
-            # so we disable CUDNN temporarily
-            original_cudnn_enabled = cudnn.enabled
-            cudnn.enabled = False
-            test_cpu_against_cuda(N, C, H, W)
-            cudnn.enabled = original_cudnn_enabled
+            if padding_mode == 'zeros':
+                groundtruth = torch.Tensor(
+                    [[0.9600, 6.0000000000, 5.0000, 4.8340, 9.0000],
+                     [2.2500, 6.333250045, 5.0000, 5.1000, 7.0000]]).view(1, 1, 2, 5)
+            else:
+                groundtruth = torch.Tensor(
+                    [[1.2000, 6.0000000000, 5.0000, 4.8340, 9.0000],
+                     [2.2500, 6.333250045, 5.0000, 5.1000, 8.7500]]).view(1, 1, 2, 5)
+
+            self.assertEqual(output.data, groundtruth)
+
+            # do gradcheck
+            N = random.randint(1, 8)
+            C = random.randint(1, 8)
+            H = random.randint(1, 8)
+            W = random.randint(1, 8)
+            input = Variable(torch.randn(N, C, H, W), requires_grad=True)
+            grid = Variable(torch.randn(N, H, W, 2), requires_grad=True)
+            self.assertTrue(gradcheck(
+                lambda inp, grid: F.grid_sample(inp, grid, padding_mode=padding_mode),
+                (input, grid)))
+
+            # test CUDA against CPU
+            if TEST_CUDA:
+                test_cpu_against_cuda(N, C, H, W, padding_mode)
 
     def test_affine_grid(self):
         # test known input on CPU
@@ -3653,6 +3707,18 @@ new_criterion_tests = [
 ]
 
 
+def kldivloss_no_reduce_test():
+    t = Variable(torch.randn(10, 10))
+    return dict(
+        fullname='KLDivLoss_no_reduce',
+        constructor=wrap_functional(
+            lambda i: F.kl_div(i, t.type_as(i), reduce=False)),
+        input_fn=lambda: torch.rand(10, 10).log(),
+        reference_fn=lambda i, _:
+            loss_reference_fns['KLDivLoss'](i, t.data.type_as(i), reduce=False),
+        pickle=False)
+
+
 def l1loss_no_reduce_test():
     t = Variable(torch.randn(2, 3, 4))
     return dict(
@@ -3811,6 +3877,7 @@ def smoothl1loss_no_reduce_test():
 
 
 new_module_tests = [
+    kldivloss_no_reduce_test(),
     l1loss_no_reduce_test(),
     mseloss_no_reduce_test(),
     nllloss_no_reduce_test(),
@@ -4553,7 +4620,7 @@ new_module_tests = [
         desc='dim'
     ),
     dict(
-        constructor=wrap_functional(F.softmax, dim=1),
+        constructor=wrap_functional(F.softmax, dim=-1),
         input_size=(2, 128),  # trigger the last-dim algo in CUDA
         fullname='softmax_lastdim',
         pickle=False,
@@ -4585,7 +4652,7 @@ new_module_tests = [
         pickle=False,
     ),
     dict(
-        constructor=wrap_functional(F.log_softmax, dim=1),
+        constructor=wrap_functional(F.log_softmax, dim=-1),
         input_size=(2, 128),  # trigger the last-dim algo in CUDA
         fullname='log_softmax_lastdim',
         pickle=False,

test/test_optim.py

@@ -61,13 +61,14 @@ class TestOptim(TestCase):
 
         self.assertLessEqual(params.data.dist(solution), initial_dist)
 
-    def _test_rosenbrock_sparse(self, constructor):
+    def _test_rosenbrock_sparse(self, constructor, sparse_only=False):
         params_t = torch.Tensor([1.5, 1.5])
 
-        params = Variable(torch.Tensor([1.5, 1.5]), requires_grad=True)
-        params_c = Variable(torch.Tensor([1.5, 1.5]), requires_grad=True)
+        params = Variable(params_t, requires_grad=True)
         optimizer = constructor([params])
-        optimizer_c = constructor([params_c])
+        if not sparse_only:
+            params_c = Variable(params_t.clone(), requires_grad=True)
+            optimizer_c = constructor([params_c])
 
         solution = torch.Tensor([1, 1])
         initial_dist = params.data.dist(solution)
@@ -99,8 +100,9 @@ class TestOptim(TestCase):
             # Do cyclic coordinate descent
             w = i % 2
             optimizer.step(functools.partial(eval, params, True, w))
-            optimizer_c.step(functools.partial(eval, params_c, False, w))
-            self.assertEqual(params.data, params_c.data)
+            if not sparse_only:
+                optimizer_c.step(functools.partial(eval, params_c, False, w))
+                self.assertEqual(params.data, params_c.data)
 
         self.assertLessEqual(params.data.dist(solution), initial_dist)
 
@@ -229,6 +231,11 @@ class TestOptim(TestCase):
                 lr=1e-3)
         )
 
+    def test_sgd_sparse(self):
+        self._test_rosenbrock_sparse(
+            lambda params: optim.SGD(params, lr=5e-3)
+        )
+
     def test_adam(self):
         self._test_rosenbrock(
             lambda params: optim.Adam(params, lr=1e-2),
@@ -247,6 +254,12 @@ class TestOptim(TestCase):
                 lr=1e-3)
         )
 
+    def test_sparse_adam(self):
+        self._test_rosenbrock_sparse(
+            lambda params: optim.SparseAdam(params, lr=4e-2),
+            True
+        )
+
     def test_adadelta(self):
         self._test_rosenbrock(
             lambda params: optim.Adadelta(params),

test/test_torch.py

@@ -71,6 +71,34 @@ class TestTorch(TestCase):
                     res2[i, j] = v1[i] * v2[j]
             self.assertEqual(res1, res2)
 
+    def test_addr(self):
+        types = {
+            'torch.DoubleTensor': 1e-8,
+            'torch.FloatTensor': 1e-4,
+        }
+
+        def run_test(m, v1, v2, m_transform=lambda x: x):
+            m = m_transform(m.clone())
+            ref = m.clone()
+            torch.addr(m, v1, v2, out=m)
+            for i in range(m.size(0)):
+                for j in range(m.size(1)):
+                    ref[i, j] += v1[i] * v2[j]
+            self.assertEqual(m, ref)
+
+        for tname, _prec in types.items():
+            for h, w in [(100, 110), (1, 20), (200, 2)]:
+                m = torch.randn(h, w).type(tname)
+                v1 = torch.randn(h).type(tname)
+                v2 = torch.randn(w).type(tname)
+                run_test(m, v1, v2)
+                # test transpose
+                run_test(m, v2, v1, lambda x: x.transpose(0, 1))
+                # test 0 strided
+                v1 = torch.randn(1).type(tname).expand(h)
+                run_test(m, v1, v2)
+                run_test(m, v2, v1, lambda x: x.transpose(0, 1))
+
     def test_addmv(self):
         types = {
             'torch.DoubleTensor': 1e-8,
@@ -408,6 +436,17 @@ class TestTorch(TestCase):
         test((10,))
         test((5, 5))
 
+    def test_all_any_empty(self):
+        x = torch.ByteTensor()
+        self.assertTrue(x.all())
+        self.assertFalse(x.any())
+
+    @unittest.skipIf(not torch.cuda.is_available(), 'no CUDA')
+    def test_all_any_empty_cuda(self):
+        x = torch.cuda.ByteTensor()
+        self.assertTrue(x.all())
+        self.assertFalse(x.any())
+
     def test_mv(self):
         m1 = torch.randn(100, 100)
         v1 = torch.randn(100)
@@ -1111,6 +1150,11 @@ class TestTorch(TestCase):
         torch.arange(0, 1, out=res2)
         self.assertEqual(res1, res2, 0)
 
+        # Check arange with only one argument
+        res1 = torch.arange(10)
+        res2 = torch.arange(0, 10)
+        self.assertEqual(res1, res2, 0)
+
         # Check arange for non-contiguous tensors.
         x = torch.zeros(2, 3)
         torch.arange(0, 4, out=x.narrow(1, 1, 2))
@@ -3643,6 +3687,11 @@ class TestTorch(TestCase):
         self.assertEqual(tensor.std(), tensor.std(unbiased=True))
         self.assertEqual(tensor.std(unbiased=False), tensor.std(0, unbiased=False)[0])
 
+    def test_var_stability(self):
+        tensor = torch.FloatTensor([2281.5, 2281.25])
+        self.assertEqual(tensor.var(0)[0], 0.03125)
+        self.assertEqual(tensor.var(), 0.03125)
+
     def test_view(self):
         tensor = torch.rand(15)
         template = torch.rand(3, 5)
@@ -3709,7 +3758,7 @@ class TestTorch(TestCase):
         self.assertEqual(result.size(), target, 'Error in repeat using result')
         result = tensor.repeat(torchSize)
         self.assertEqual(result.size(), target, 'Error in repeat using result and LongStorage')
-        self.assertEqual((result.mean(0).view(8, 4) - tensor).abs().max(), 0, 'Error in repeat (not equal)')
+        self.assertEqual(result.mean(0).view(8, 4), tensor, 'Error in repeat (not equal)')
 
     def test_is_same_size(self):
         t1 = torch.Tensor(3, 4, 9, 10)
@@ -4511,6 +4560,19 @@ class TestTorch(TestCase):
             for i in range(len(x)):
                 self.assertEqual(geq2_x[i], geq2_array[i])
 
+    def test_error_msg_type_translation(self):
+        with self.assertRaisesRegex(
+                RuntimeError,
+                # message includes both torch.DoubleTensor and torch.LongTensor
+                '(?=.*torch\.DoubleTensor)(?=.*torch\.LongTensor)'):
+
+            # Calls model with a DoubleTensor input but LongTensor weights
+            input = torch.autograd.Variable(torch.randn(1, 1, 1, 6).double())
+            weight = torch.zeros(1, 1, 1, 3).long()
+            model = torch.nn.Conv2d(1, 1, (1, 3), stride=1, padding=0, bias=False)
+            model.weight.data = weight
+            out = model(input)
+
     def test_comparison_ops(self):
         x = torch.randn(5, 5)
         y = torch.randn(5, 5)

test/test_utils.py

@@ -386,7 +386,7 @@ class TestONNXUtils(TestCase):
         sizes = [2, 3, 4]
         pad = [1, 2, 3, 4]
         paddings = prepare_onnx_paddings(len(sizes), pad)
-        self.assertEqual(paddings, [0, 0, 3, 4, 1, 2])
+        self.assertEqual(paddings, [0, 3, 1, 0, 4, 2])
 
     def test_check_onnx_broadcast(self):
 

tools/autograd/derivatives.yaml

@@ -13,10 +13,10 @@
 
 - name: add(Tensor self, Tensor other, *, Scalar alpha=1)
   self: grad
-  other: grad * alpha
+  other: maybe_multiply(grad, alpha)
 
 - name: addbmm(Tensor self, Tensor batch1, Tensor batch2, *, Scalar beta=1, Scalar alpha=1)
-  self: grad * beta
+  self: maybe_multiply(grad, beta)
   batch1: grad.unsqueeze(0).expand({ batch1.size(0), batch1.size(1), batch2.size(2) }).bmm(batch2.transpose(1, 2)) * alpha
   batch2: batch1.transpose(1, 2).bmm(grad.unsqueeze(0).expand({ batch1.size(0), batch1.size(1), batch2.size(2) })) * alpha
 
@@ -36,12 +36,12 @@
   mat2: mm_mat2_backward(grad, mat1, mat2.sizes(), mat2.strides(), alpha)
 
 - name: addmv(Tensor self, Tensor mat, Tensor vec, *, Scalar beta=1, Scalar alpha=1)
-  self: grad * beta
+  self: maybe_multiply(grad, beta)
   mat: grad.ger(vec) * alpha
   vec: mat.t().mv(grad) * alpha
 
 - name: addr(Tensor self, Tensor vec1, Tensor vec2, *, Scalar beta=1, Scalar alpha=1)
-  self: grad * beta
+  self: maybe_multiply(grad, beta)
   vec1: grad.mv(vec2) * alpha
   vec2: grad.t().mv(vec1) * alpha
 
@@ -62,7 +62,7 @@
   other: grad * -self * ((self * self + other * other).reciprocal())
 
 - name: baddbmm(Tensor self, Tensor batch1, Tensor batch2, *, Scalar beta=1, Scalar alpha=1)
-  self: grad * beta
+  self: maybe_multiply(grad, beta)
   batch1: grad.bmm(batch2.transpose(1, 2)) * alpha
   batch2: batch1.transpose(1, 2).bmm(grad) * alpha
 
@@ -108,8 +108,8 @@
   self: grad.diag(diagonal)
 
 - name: dist(Tensor self, Tensor other, Scalar p=2)
-  self: norm_backward(grad, self - other, p)
-  other: -norm_backward(grad, self - other, p)
+  self: norm_backward(grad, self - other, p, result)
+  other: -norm_backward(grad, self - other, p, result)
 
 - name: div(Tensor self, Scalar other)
   self: grad / other
@@ -149,7 +149,8 @@
 
 - name: eye  # fallthrough
 
-- name: fill(Tensor self, Scalar value)  # FIXME
+- name: fill(Tensor self, Scalar value)
+  self: zeros_like(grad)
 
 - name: floor(Tensor self)
   self: zeros_like(grad)
@@ -217,7 +218,6 @@
 
 - name: index_select(Tensor self, int64_t dim, Tensor index)
   self: grad.type().zeros(self.sizes()).index_add_(dim, index, grad)
-  __view__: True
 
 - name: inverse(Tensor self)
   self: -at::mm(output.t(), at::mm(grad, output.t()))
@@ -348,10 +348,10 @@
   self: zeros_like(grad)
 
 - name: norm(Tensor self, Scalar p=2)
-  self: norm_backward(grad, self, p)
+  self: norm_backward(grad, self, p, result)
 
 - name: norm(Tensor self, Scalar p, int64_t dim, bool keepdim=False)
-  self: norm_backward(grad, self, p, dim, keepdim)
+  self: norm_backward(grad, self, p, destination, dim, keepdim)
 
 - name: numel  # fallthrough
 - name: ones  # fallthrough
@@ -395,7 +395,7 @@
   self: not_implemented("pstrf")
 
 - name: put(Tensor self, Tensor index, Tensor source, bool accumulate)
-  self: zeros_like(self).put_(index, source, accumulate)
+  self: grad.clone().put_(index, zeros_like(source), accumulate)
   source: grad.take(index)
 
 - name: qr(Tensor self)
@@ -468,7 +468,7 @@
   __view__: True
 
 - name: squeeze(Tensor self, int64_t dim)
-  self: maybe_unsqueeze(grad, dim, self.size(dim) == 1)
+  self: maybe_unsqueeze(grad, dim, self.size(dim) == 1 && self.sizes().size() != 1)
   __view__: True
 
 - name: std
@@ -563,9 +563,9 @@
   grad_output: avg_pool3d(grad, kernel_size, stride, padding, ceil_mode, count_include_pad)
   input: zeros_like(input)
 
-- name: elu_backward(Tensor grad_output, Tensor input, Scalar alpha, bool inplace, Tensor output)
-  grad_output: elu_backward(grad, input, alpha, inplace, output)
-  input: grad * grad_input * (input < 0).toType(grad.type())
+- name: elu_backward(Tensor grad_output, Scalar alpha, Tensor output)
+  grad_output: elu_backward(grad, alpha, output)
+  output: grad * grad_output * (output < 0).toType(grad.type())
 
 - name: glu_backward(Tensor grad_output, Tensor input, int64_t dim)
   grad_output: glu_double_backward_grad_output(grad, input, dim)
@@ -575,11 +575,12 @@
   grad_output: hardshrink_backward(grad, input, lambd)
   input: zeros_like(grad)
 
-- name: hardtanh_backward(Tensor grad_output, Tensor input, Scalar min_val, Scalar max_val, bool inplace)
-  grad_output: hardtanh_backward(grad, input, min_val, max_val, false)
+- name: hardtanh_backward(Tensor grad_output, Tensor input, Scalar min_val, Scalar max_val)
+  grad_output: hardtanh_backward(grad, input, min_val, max_val)
   input: zeros_like(grad)
 
-- name: kl_div_backward(Tensor input, Tensor target, bool size_average)
+- name: kl_div_backward(Tensor grad_output, Tensor input, Tensor target, bool size_average, bool reduce)
+  grad_output: kl_div_double_backward_grad_output(grad, input, target, size_average, reduce)
   input: zeros_like(grad)
 
 - name: l1_loss_backward(Tensor grad_output, Tensor input, Tensor target, bool size_average, bool reduce)
@@ -594,8 +595,8 @@
   grad_output: grad - (grad * output.exp()).sum(dim, true)
   input: log_softmax_double_backward(grad, grad_output, dim, output)
 
-- name: leaky_relu_backward(Tensor grad_output, Tensor input, Scalar negative_slope, bool inplace)
-  grad_output: leaky_relu_backward(grad, input, negative_slope, false)
+- name: leaky_relu_backward(Tensor grad_output, Tensor input, Scalar negative_slope)
+  grad_output: leaky_relu_backward(grad, input, negative_slope)
   input: zeros_like(grad)
 
 - name: max_pool2d_backward(Tensor grad_output, Tensor input, IntList kernel_size, IntList stride, IntList padding, IntList dilation, bool ceil_mode, Tensor indices)
@@ -623,8 +624,8 @@
   input: zeros_like(input)
   weight: zeros_like(weight)
 
-- name: rrelu_backward(Tensor grad_output, Tensor input, Scalar lower, Scalar upper, bool training, bool inplace, Tensor noise)
-  grad_output: rrelu_backward(grad, input, lower, upper, training, false, noise)
+- name: rrelu_backward(Tensor grad_output, Tensor input, Scalar lower, Scalar upper, bool training, Tensor noise)
+  grad_output: rrelu_backward(grad, input, lower, upper, training, noise)
   input: zeros_like(grad)
 
 - name: smooth_l1_loss_backward(Tensor grad_output, Tensor input, Tensor target, bool size_average, bool reduce)
@@ -646,8 +647,8 @@
   grad_output: softshrink_backward(grad, input, lambd)
   input: zeros_like(grad)
 
-- name: threshold_backward(Tensor grad_output, Tensor input, Scalar threshold, Scalar value, bool inplace)
-  grad_output: threshold_backward(grad, input, threshold, value, false)
+- name: threshold_backward(Tensor grad_output, Tensor input, Scalar threshold, Scalar value)
+  grad_output: threshold_backward(grad, input, threshold, value)
   input: zeros_like(grad)
 
 - name: _sigmoid_backward(Tensor grad_output, Tensor output)

tools/autograd/gen_python_functions.py

@@ -49,6 +49,16 @@ PY_VARIABLE_METHOD_DEF = CodeTemplate("""\
 UNPACK_SELF = "auto& self_ = reinterpret_cast<THPVariable*>(self)->cdata;"
 
 
+# XXX: if you got here because of an assertion failure, it doesn't mean
+# it's enough to just extend the list here. Before you do this, make sure
+# to add an appropriate wrap() overload in torch/csrc/autograd/utils/wrap_outputs.h.
+SUPPORTED_RETURN_TYPES = {
+    'Tensor', 'std::tuple<Tensor,Tensor>',
+    'std::tuple<Tensor,Tensor,Tensor>', 'std::vector<Tensor>',
+    'Scalar', 'bool', 'int64_t', 'void*'
+}
+
+
 def create_python_bindings(
         python_functions, py_methods, py_method_defs, py_method_dispatch,
         is_class):
@@ -80,6 +90,9 @@ def create_python_bindings(
 
     def emit_dispatch(i, function):
         env = {}
+        simple_return_type = function['return_type'].replace(' &', '')
+        assert simple_return_type in SUPPORTED_RETURN_TYPES, \
+            function['name'] + ' returns unsupported type: ' + simple_return_type
 
         actuals = []
         formal_args = []

tools/autograd/gen_variable_type.py

@@ -39,7 +39,11 @@ return baseType->${method_prefix}${api_name}(${unpacked_args});""")
 
 METHOD_DEFINITION_FALLTHROUGH_VARIABLE = CodeTemplate("""\
 ${unpack_args}
-return as_variable(baseType->${method_prefix}${api_name}(${unpacked_args}));""")
+auto flags = compute_flags({ ${args_with_derivatives} });
+auto var = as_variable(baseType->${method_prefix}${api_name}(${unpacked_args}));
+var.is_volatile() = flags.is_volatile;
+return var;
+""")
 
 METHOD_DEFINITION_FALLTHROUGH_INPLACE = CodeTemplate("""\
 ${unpack_args}
@@ -67,6 +71,7 @@ FUNCTION_DEFINITION = CodeTemplate("""\
 variable_list ${op}::apply(const variable_list& grads) {
   variable_list grad_inputs{${num_inputs}};
   ${body}
+  ensure_no_aten_scalars(grad_inputs);
   return grad_inputs;
 }
 """)
@@ -682,11 +687,6 @@ def create_variable_type(top_env, aten_declarations):
         if declaration['return_type'] in FALLTHROUGH_RETURN_TYPES:
             body.extend(METHOD_DEFINITION_FALLTHROUGH.substitute(combined).split('\n'))
             return body
-        elif declaration['name'] in FALLTHROUGH_FUNCTIONS:
-            tmpl = (METHOD_DEFINITION_FALLTHROUGH_INPLACE if declaration['inplace']
-                    else METHOD_DEFINITION_FALLTHROUGH_VARIABLE)
-            body.extend(tmpl.substitute(combined).split('\n'))
-            return body
 
         arguments = declaration['arguments']
         tensor_args = [arg for arg in arguments if arg['simple_type'] in {'Tensor', 'TensorList'}]
@@ -752,6 +752,12 @@ def create_variable_type(top_env, aten_declarations):
         elif is_view:
             env['version_counter'] = 'take_version_counter(ret, self);'
 
+        if declaration['name'] in FALLTHROUGH_FUNCTIONS:
+            tmpl = (METHOD_DEFINITION_FALLTHROUGH_INPLACE if declaration['inplace']
+                    else METHOD_DEFINITION_FALLTHROUGH_VARIABLE)
+            body.extend(tmpl.substitute(combined).split('\n'))
+            return body
+
         base_call = BASE_CALL.substitute(combined)
         if not declaration['inplace']:
             base_call = 'auto ret = as_variable({})'.format(base_call)

tools/autograd/templates/Functions.cpp

@@ -34,41 +34,44 @@ Tensor maybe_multiply(const Tensor & t, const Scalar & s) {
   }
 }
 
-Tensor norm_backward(const Tensor & grad, const Tensor & self, const Scalar & p_) {
-  auto p = p_.toDouble();
-  auto norm = self.norm(p_);
-
-  if (norm.toDouble() == 0.0) {
-    // handle case at 0 where we return a subgradient containing 0
-    return zeros_like(self);
-  }
-
-  if (p == 2.0) {
-    return self * (grad / norm);
-  } else {
-    auto pow_ = self.abs().pow(p - 2);
-    auto scale_v = grad / norm.toTensor().pow(p - 1);
-    return self * pow_ * scale_v;
+// Don't expose ATen scalars to Variable API, because they are not supported yet.
+void ensure_no_aten_scalars(variable_list &vars) {
+  for (auto& v : vars) {
+    if (v.defined() && v.dim() == 0) {
+      v.data().as_strided_({1}, {1});
+    }
   }
 }
 
-Tensor norm_backward(Tensor grad, const Tensor & self, const Scalar & p_, int64_t dim, bool keepdim) {
-  if (!keepdim && self.dim() > 1) {
-    grad = grad.unsqueeze(dim);
-  }
-  auto p = p_.toDouble();
-  auto norm = self.norm(p, dim, true);
-  Tensor grad_input;
-  if (p == 2.0) {
-    grad_input = self * (grad / norm);
+Tensor norm_backward(const Tensor & grad, const Tensor & self, const Scalar & p_, const Tensor & norm) {
+  double p = p_.toDouble();
+  Tensor self_scaled;
+  Tensor scale_v;
+  if (p == 0.0) {
+    return zeros_like(self);
+  } else if (p == 1.0) {
+    return self.sign() * grad;
+  } else if (p < 2.0) {
+    self_scaled = self.sign() * self.abs().pow(p - 1);
+    scale_v = grad / norm.pow(p - 1);
+  } else if (p == 2.0) {
+    self_scaled = self;
+    scale_v = grad / norm;
   } else {
-    auto pow_ = self.abs().pow(p - 2);
-    auto scale_v = grad / norm.pow(p - 1);
-    grad_input = self * pow_ * scale_v;
+    self_scaled = self * self.abs().pow(p - 2);
+    scale_v = grad / norm.pow(p - 1);
   }
   // handle case at 0 where we return a subgradient containing 0
-  grad_input.masked_fill_(norm == 0, 0);
-  return grad_input;
+  scale_v.masked_fill_(norm == 0, 0);
+  return self_scaled * scale_v;
+}
+
+Tensor norm_backward(Tensor grad, const Tensor & self, const Scalar & p_, Tensor norm, int64_t dim, bool keepdim) {
+  if (!keepdim && self.dim() > 1) {
+    grad = grad.unsqueeze(dim);
+    norm = norm.unsqueeze(dim);
+  }
+  return norm_backward(grad, self, p_, norm);
 }
 
 Tensor reduce_to(const Tensor & grad, IntList sizes) {
@@ -300,6 +303,16 @@ Tensor glu_double_backward_grad_output(const Tensor & grad, const Tensor & input
   return tmp.narrow(dim, 0, sizes[dim]) + tmp.narrow(dim, sizes[dim], sizes[dim]);
 }
 
+Tensor kl_div_double_backward_grad_output(const Tensor & grad, const Tensor & input, const Tensor & target, bool size_average, bool reduce) {
+  auto result = kl_div_backward(grad, input, target, size_average, false);
+  if (reduce && size_average) {
+    return result.mean().toTensor();
+  } else if (reduce) {
+    return result.sum().toTensor();
+  }
+  return result;
+}
+
 Tensor log_sigmoid_double_backward(const Tensor & grad, const Tensor & input) {
   auto z = input.sigmoid();
   return grad * (z - 1) * z;

tools/docker/Dockerfile9

@@ -25,7 +25,7 @@ RUN curl -o ~/miniconda.sh -O  https://repo.continuum.io/miniconda/Miniconda3-la
      /opt/conda/bin/conda create -y --name pytorch-py$PYTHON_VERSION python=$PYTHON_VERSION numpy pyyaml scipy ipython mkl&& \
      /opt/conda/bin/conda clean -ya 
 ENV PATH /opt/conda/envs/pytorch-py$PYTHON_VERSION/bin:$PATH
-#RUN conda install --name pytorch-py$PYTHON_VERSION -c soumith magma-cuda80
+RUN conda install --name pytorch-py$PYTHON_VERSION -c soumith magma-cuda90
 # This must be done before pip so that requirements.txt is available
 WORKDIR /opt/pytorch
 COPY . .

tools/pytorch.version

@@ -0,0 +1,31 @@
+{
+     global:
+         _TH*;
+         __TH*;
+         TH*;
+         *THP*;
+         *THCP*;
+         PyInit*;
+         init*;
+         state;
+	 _ZGVZN2at*;
+         _ZN2at*;
+	 _ZNK2at*Type*;
+	 _ZNK2at*Tensor*;
+	 _ZNK2at*Storage*;
+	 _ZNK2at*Scalar*;
+	 _ZNK2at*CUDA*;
+	 *2at7Context*;
+	 _ZTIN2at*;
+	 _ZTIZN2at*;
+	 _ZTSN2at*;
+	 _ZTSPN2at*;
+	 _ZTSZN2at*;
+	 _ZTVN2at*;
+	 _ZZN2at*;
+	 _Z*torch*;
+	 _Z*Tensor*;
+	 _Z*tensor*;
+     local:
+         *;
+ };

torch/_tensor_docs.py

@@ -322,10 +322,10 @@ It may be of a different data type or reside on a different device.
 
 Args:
     src (Tensor): Source tensor to copy
-    async (bool): If True and this copy is between CPU and GPU, then the copy
+    async (bool): If ``True`` and this copy is between CPU and GPU, then the copy
         may occur asynchronously with respect to the host. For other
         copies, this argument has no effect.
-    broadcast (bool): If True, :attr:`src` will be broadcast to the shape of
+    broadcast (bool): If ``True``, :attr:`src` will be broadcast to the shape of
         the underlying tensor.
 """)
 

torch/_torch_docs.py

@@ -1244,7 +1244,7 @@ Computes the eigenvalues and eigenvectors of a real square matrix.
 Args:
     a (Tensor): A square matrix for which the eigenvalues and eigenvectors will
                 be computed
-    eigenvectors (bool): `True` to compute both eigenvalues and eigenvectors.
+    eigenvectors (bool): ``True`` to compute both eigenvalues and eigenvectors.
                          Otherwise, only eigenvalues will be computed.
     out (tuple, optional): Output tensors
 
@@ -1287,7 +1287,7 @@ add_docstr(torch._C.equal,
            """
 equal(tensor1, tensor2) -> bool
 
-True if two tensors have the same size and elements, False otherwise.
+``True`` if two tensors have the same size and elements, ``False`` otherwise.
 
 Example::
 
@@ -1843,7 +1843,7 @@ If :attr:`dim` is not given, the last dimension of the `input` is chosen.
 A tuple of `(values, indices)` is returned, where the `indices` is the indices
 of the kth-smallest element in the original `input` Tensor in dimension `dim`.
 
-If :attr:`keepdim` is true, both the :attr:`values` and :attr:`indices` Tensors
+If :attr:`keepdim` is ``True``, both the :attr:`values` and :attr:`indices` Tensors
 are the same size as :attr:`input`, except in the dimension :attr:`dim` where
 they are of size 1. Otherwise, :attr:`dim` is squeezed
 (see :func:`torch.squeeze`), resulting in both the :attr:`values` and
@@ -2230,7 +2230,7 @@ Returns the maximum value of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`. The second return value is the index location of each
 maximum value found (argmax).
 
-If :attr:`keepdim` is true, the output Tensors are of the same size
+If :attr:`keepdim` is ``True``, the output Tensors are of the same size
 as :attr:`input` except in the dimension :attr:`dim` where they are of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting
 in the output Tensors having 1 fewer dimension than :attr:`input`.
@@ -2341,7 +2341,7 @@ Example::
 Returns the mean value of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensor is of the same size
+If :attr:`keepdim` is ``True``, the output Tensor is of the same size
 as :attr:`input` except in the dimension :attr:`dim` where it is of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting in the
 output Tensor having 1 fewer dimension.
@@ -2411,7 +2411,7 @@ as a `LongTensor`.
 
 By default, :attr:`dim` is the last dimension of the :attr:`input` Tensor.
 
-If :attr:`keepdim` is true, the output Tensors are of the same size
+If :attr:`keepdim` is ``True``, the output Tensors are of the same size
 as :attr:`input` except in the dimension :attr:`dim` where they are of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting in
 the outputs Tensor having 1 fewer dimension than :attr:`input`.
@@ -2486,7 +2486,7 @@ Returns the minimum value of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`. The second return value is the index location of each
 minimum value found (argmin).
 
-If :attr:`keepdim` is true, the output Tensors are of the same size as
+If :attr:`keepdim` is ``True``, the output Tensors are of the same size as
 :attr:`input` except in the dimension :attr:`dim` where they are of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting in
 the output Tensors having 1 fewer dimension than :attr:`input`.
@@ -2608,7 +2608,7 @@ as a `LongTensor`.
 
 By default, :attr:`dim` is the last dimension of the :attr:`input` Tensor.
 
-If :attr:`keepdim` is true, the output Tensors are of the same size as
+If :attr:`keepdim` is ``True``, the output Tensors are of the same size as
 :attr:`input` except in the dimension :attr:`dim` where they are of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting
 in the output Tensors having 1 fewer dimension than :attr:`input`.
@@ -2756,7 +2756,7 @@ If :attr:`input` is a vector, :attr:`out` is a vector of size `num_samples`.
 If :attr:`input` is a matrix with `m` rows, :attr:`out` is an matrix of shape
 `m \u00D7 n`.
 
-If replacement is `True`, samples are drawn with replacement.
+If replacement is ``True``, samples are drawn with replacement.
 
 If not, they are drawn without replacement, which means that when a
 sample index is drawn for a row, it cannot be drawn again for that row.
@@ -2945,7 +2945,7 @@ Example::
 Returns the p-norm of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensor is of the same size as
+If :attr:`keepdim` is ``True``, the output Tensor is of the same size as
 :attr:`input` except in the dimension :attr:`dim` where it is of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting
 in the output Tensor having 1 fewer dimension than :attr:`input`.
@@ -3156,9 +3156,9 @@ potrf(a, upper, out=None)
 
 Computes the Cholesky decomposition of a positive semidefinite
 matrix :attr:`a`: returns matrix `u`
-If `upper` is True or not provided, `u` is upper triangular
+If `upper` is ``True`` or not provided, `u` is upper triangular
 such that :math:`a = u^T u`.
-If `upper` is False, `u` is lower triangular
+If `upper` is ``False``, `u` is lower triangular
 such that :math:`a = u u^T`.
 
 Args:
@@ -3201,9 +3201,9 @@ potri(u, upper, out=None)
 
 Computes the inverse of a positive semidefinite matrix given its
 Cholesky factor :attr:`u`: returns matrix `inv`
-If `upper` is True or not provided, `u` is upper triangular
+If `upper` is ``True`` or not provided, `u` is upper triangular
 such that :math:`inv = (u^T u)^{-1}`.
-If `upper` is False, `u` is lower triangular
+If `upper` is ``False``, `u` is lower triangular
 such that :math:`inv = (u u^T)^{-1}`.
 
 Args:
@@ -3248,9 +3248,9 @@ potrs(b, u, upper, out=None)
 Solves a linear system of equations with a positive semidefinite
 matrix to be inverted given its given a Cholesky factor
 matrix :attr:`u`: returns matrix `c`
-If `upper` is True or not provided, `u` is and upper triangular
+If `upper` is ``True`` or not provided, `u` is and upper triangular
 such that :math:`c = (u^T u)^{-1} b`.
-If `upper` is False, `u` is and lower triangular
+If `upper` is ``False``, `u` is and lower triangular
 such that :math:`c = (u u^T)^{-1} b`.
 
 .. note:: `b` is always a 2D `Tensor`, use `b.unsqueeze(1)` to convert a vector.
@@ -3424,7 +3424,7 @@ Example::
 Returns the product of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensor is of the same size as
+If :attr:`keepdim` is ``True``, the output Tensor is of the same size as
 :attr:`input` except in the dimension :attr:`dim` where it is of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting
 in the output Tensor having 1 fewer dimension than :attr:`input`.
@@ -3463,9 +3463,9 @@ pstrf(a, upper, out=None)
 
 Computes the pivoted Cholesky decomposition of a positive semidefinite
 matrix :attr:`a`: returns matrices `u` and `piv`.
-If `upper` is True or not provided, `u` is and upper triangular
+If `upper` is ``True`` or not provided, `u` is and upper triangular
 such that :math:`a = p^T u^T u p`, with `p` the permutation given by `piv`.
-If `upper` is False, `u` is and lower triangular
+If `upper` is ``False``, `u` is and lower triangular
 such that :math:`a = p^T u u^T p`.
 
 Args:
@@ -3691,7 +3691,7 @@ Example::
 
 add_docstr(torch._C.arange,
            """
-arange(start, end, step=1, out=None) -> Tensor
+arange(start=0, end, step=1, out=None) -> Tensor
 
 Returns a 1D Tensor of size :math:`floor((end - start) / step)` with values
 from the interval ``[start, end)`` taken with step :attr:`step` starting
@@ -3705,6 +3705,15 @@ Args:
 
 Example::
 
+    >>> torch.arange(5)
+
+     0
+     1
+     2
+     3
+     4
+    [torch.FloatTensor of size 5]
+
     >>> torch.arange(1, 4)
 
      1
@@ -3989,7 +3998,7 @@ in ascending order by value.
 
 If :attr:`dim` is not given, the last dimension of the `input` is chosen.
 
-If :attr:`descending` is `True` then the elements are sorted in descending
+If :attr:`descending` is ``True`` then the elements are sorted in descending
 order by value.
 
 A tuple of (sorted_tensor, sorted_indices) is returned, where the
@@ -4117,7 +4126,7 @@ add_docstr(torch._C.std,
 
 Returns the standard-deviation of all elements in the :attr:`input` Tensor.
 
-If :attr:`unbiased` is false, then the standard-deviation will be calculated via
+If :attr:`unbiased` is ``False``, then the standard-deviation will be calculated via
 the biased estimator. Otherwise, Bessel's correction will be used.
 
 Args:
@@ -4141,12 +4150,12 @@ Example::
 Returns the standard-deviation of each row of the :attr:`input` Tensor in the
 given dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensor is of the same size as
+If :attr:`keepdim` is ``True``, the output Tensor is of the same size as
 :attr:`input` except in the dimension :attr:`dim` where it is of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting
 in the output Tensor having 1 fewer dimension than :attr:`input`.
 
-If :attr:`unbiased` is false, then the standard-deviation will be calculated via
+If :attr:`unbiased` is ``False``, then the standard-deviation will be calculated via
 the biased estimator. Otherwise, Bessel's correction will be used.
 
 Args:
@@ -4203,7 +4212,7 @@ Example::
 Returns the sum of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensor is of the same size
+If :attr:`keepdim` is ``True``, the output Tensor is of the same size
 as :attr:`input` except in the dimension :attr:`dim` where it is of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting in
 the output Tensor having 1 fewer dimension than :attr:`input`.
@@ -4325,13 +4334,13 @@ such that `input = V diag(e) V'`
 The boolean argument :attr:`eigenvectors` defines computation of
 eigenvectors or eigenvalues only.
 
-If it is `False`, only eigenvalues are computed. If it is `True`,
+If it is ``False``, only eigenvalues are computed. If it is ``True``,
 both eigenvalues and eigenvectors are computed.
 
 Since the input matrix `input` is supposed to be symmetric,
 only the upper triangular portion is used by default.
 
-If :attr:`upper` is `False`, then lower triangular portion is used.
+If :attr:`upper` is ``False``, then lower triangular portion is used.
 
 Note: Irrespective of the original strides, the returned matrix `V` will
 be transposed, i.e. with strides `(1, m)` instead of `(m, 1)`.
@@ -4493,12 +4502,12 @@ a given dimension.
 
 If :attr:`dim` is not given, the last dimension of the `input` is chosen.
 
-If :attr:`largest` is `False` then the `k` smallest elements are returned.
+If :attr:`largest` is ``False`` then the `k` smallest elements are returned.
 
 A tuple of `(values, indices)` is returned, where the `indices` are the indices
 of the elements in the original `input` Tensor.
 
-The boolean option :attr:`sorted` if `True`, will make sure that the returned
+The boolean option :attr:`sorted` if ``True``, will make sure that the returned
 `k` elements are themselves sorted
 
 Args:
@@ -4787,7 +4796,7 @@ add_docstr(torch._C.var,
 
 Returns the variance of all elements in the :attr:`input` Tensor.
 
-If :attr:`unbiased` is false, then the variance will be calculated via the
+If :attr:`unbiased` is ``False``, then the variance will be calculated via the
 biased estimator. Otherwise, Bessel's correction will be used.
 
 Args:
@@ -4811,12 +4820,12 @@ Example::
 Returns the variance of each row of the :attr:`input` Tensor in the given
 dimension :attr:`dim`.
 
-If :attr:`keepdim` is true, the output Tensors are of the same size
+If :attr:`keepdim` is ``True``, the output Tensors are of the same size
 as :attr:`input` except in the dimension :attr:`dim` where they are of size 1.
 Otherwise, :attr:`dim` is squeezed (see :func:`torch.squeeze`), resulting in
 the outputs Tensor having 1 fewer dimension than :attr:`input`.
 
-If :attr:`unbiased` is false, then the variance will be calculated via the
+If :attr:`unbiased` is ``False``, then the variance will be calculated via the
 biased estimator. Otherwise, Bessel's correction will be used.
 
 Args:

torch/_utils.py

@@ -12,7 +12,7 @@ def _type(self, new_type=None, async=False):
 
     Args:
         new_type (type or string): The desired type
-        async (bool): If True, and the source is in pinned memory and
+        async (bool): If ``True``, and the source is in pinned memory and
                       destination is on the GPU or vice versa, the copy is
                       performed asynchronously with respect to the host.
                       Otherwise, the argument has no effect.
@@ -46,7 +46,7 @@ def _cuda(self, device=None, async=False):
 
     Args:
         device (int): The destination GPU id. Defaults to the current device.
-        async (bool): If True and the source is in pinned memory, the copy will
+        async (bool): If ``True`` and the source is in pinned memory, the copy will
                       be asynchronous with respect to the host. Otherwise, the
                       argument has no effect.
     """

torch/autograd/__init__.py

@@ -63,16 +63,16 @@ def backward(variables, grad_variables=None, retain_graph=None, create_graph=Non
         grad_variables (sequence of (Tensor, Variable or None)): Gradients w.r.t.
             each element of corresponding variables.  Any tensors will be
             automatically converted to Variables that are volatile unless
-            ``create_graph`` is True.  None values can be specified for scalar
+            ``create_graph`` is ``True``.  None values can be specified for scalar
             Variables or ones that don't require grad. If a None value would
             be acceptable for all grad_variables, then this argument is optional.
-        retain_graph (bool, optional): If False, the graph used to compute the grad
-            will be freed. Note that in nearly all cases setting this option to True
+        retain_graph (bool, optional): If ``False``, the graph used to compute the grad
+            will be freed. Note that in nearly all cases setting this option to ``True``
             is not needed and often can be worked around in a much more efficient
             way. Defaults to the value of ``create_graph``.
-        create_graph (bool, optional): If true, graph of the derivative will
+        create_graph (bool, optional): If ``True``, graph of the derivative will
             be constructed, allowing to compute higher order derivative products.
-            Defaults to False, unless ``grad_variables`` contains at least one
+            Defaults to ``False``, unless ``grad_variables`` contains at least one
             non-volatile Variable.
     """
     variables = (variables,) if isinstance(variables, Variable) else tuple(variables)
@@ -109,8 +109,8 @@ def grad(outputs, inputs, grad_outputs=None, retain_graph=None, create_graph=Non
     Gradients can be given as Tensors when one doesn't need the graph of the
     derivative, or as Variables, in which case the graph will be created.
 
-    If ``only_inputs`` is True, the function will only return a list of gradients
-    w.r.t the specified inputs. If it's False, then gradient w.r.t. all remaining
+    If ``only_inputs`` is ``True``, the function will only return a list of gradients
+    w.r.t the specified inputs. If it's ``False``, then gradient w.r.t. all remaining
     leaves will still be computed, and will be accumulated into their ``.grad``
     attribute.
 
@@ -120,24 +120,24 @@ def grad(outputs, inputs, grad_outputs=None, retain_graph=None, create_graph=Non
             returned (and not accumulated into ``.grad``).
         grad_outputs (sequence of Tensor or Variable): Gradients w.r.t. each output.
             Any tensors will be automatically converted to Variables that are
-            volatile unless ``create_graph`` is True.  None values can be
+            volatile unless ``create_graph`` is ``True``. None values can be
             specified for scalar Variables or ones that don't require grad.
             If a None value would be acceptable for all grad_variables, then
             this argument is optional.
-        retain_graph (bool, optional): If False, the graph used to compute the grad
-            will be freed. Note that in nearly all cases setting this option to True
+        retain_graph (bool, optional): If ``False``, the graph used to compute the grad
+            will be freed. Note that in nearly all cases setting this option to ``True``
             is not needed and often can be worked around in a much more efficient
             way. Defaults to the value of ``create_graph``.
-        create_graph (bool, optional): If True, graph of the derivative will
+        create_graph (bool, optional): If ``True``, graph of the derivative will
             be constructed, allowing to compute higher order derivative products.
-            Defaults to False, unless ``grad_variables`` contains at least one
+            Defaults to ``False``, unless ``grad_variables`` contains at least one
             non-volatile Variable.
-        only_inputs (bool, optional): If True, gradient w.r.t. leaves that are
+        only_inputs (bool, optional): If ``True``, gradient w.r.t. leaves that are
             part of the graph, but don't appear in ``inputs`` won't be computed
-            and accumulated. Defaults to True.
-        allow_unused (bool, optional): If False, specifying inputs that were not
+            and accumulated. Defaults to ``True``.
+        allow_unused (bool, optional): If ``False``, specifying inputs that were not
             used when computing outputs (and therefore their grad is always zero)
-            is an error. Default: False.
+            is an error. Defaults to ``False``.
     """
 
     outputs = (outputs,) if isinstance(outputs, Variable) else tuple(outputs)

torch/autograd/_functions/stochastic.py

@@ -2,7 +2,7 @@ import torch
 from ..function import Function
 
 
-class Multinomial(Function):
+class Categorical(Function):
     @staticmethod
     def forward(ctx, probs, num_samples, with_replacement):
         samples = probs.multinomial(num_samples, with_replacement)

torch/autograd/_functions/utils.py

@@ -57,15 +57,14 @@ def maybe_unexpand_or_view(variable, old_size):
 #          The order is dim_n_begin, dim_n_end, dim_n-1_begin, dim_n-1_end, ...
 def prepare_onnx_paddings(dim, pad):
     assert isinstance(dim, int)
-    # The order of paddings is dim_0_begin, dim_0_end, dim_1_begin, ... , dim_n_end.
+    # The desired order of paddings is
+    # dim_0_begin, dim_1_begin, ... , dim_0_end, ..., dim_n_end.
     # n is the dimension of input.
     assert len(pad) <= dim * 2
-    paddings = []
-    # pad is guaranteed to have even elements.
-    for i, j in zip(pad[0::2], pad[1::2]):
-        paddings = [i, j] + paddings
-    while len(paddings) < 2 * dim:
-        paddings = [0, 0] + paddings
+    # assume zero-dimensions in the beginning
+    paddings = list(pad[:]) + [0] * (dim * 2 - len(pad))
+    # reverse order and collate first beginnings and then ends
+    paddings = paddings[-2::-2] + paddings[-1::-2]
     assert len(paddings) == dim * 2
     return paddings
 

torch/autograd/gradcheck.py

@@ -203,7 +203,7 @@ def gradcheck(func, inputs, eps=1e-6, atol=1e-5, rtol=1e-3, raise_exception=True
     return True
 
 
-def gradgradcheck(func, inputs, grad_outputs, eps=1e-6, atol=1e-5, rtol=1e-3):
+def gradgradcheck(func, inputs, grad_outputs=None, eps=1e-6, atol=1e-5, rtol=1e-3):
     """Check gradients of gradients computed via small finite differences
        against analytical gradients
     This function checks that backpropagating through the gradients computed
@@ -216,17 +216,27 @@ def gradgradcheck(func, inputs, grad_outputs, eps=1e-6, atol=1e-5, rtol=1e-3):
     is true for all elements of analytical gradient a and numerical gradient n.
 
     Args:
-        func: Python function that takes Variable inputs and returns
+        func (function): Python function that takes Variable inputs and returns
             a tuple of Variables
-        inputs: tuple of Variables
-        grad_outputs: tuple of Variables
-        eps: perturbation for finite differences
-        atol: absolute tolerance
-        rtol: relative tolerance
+        inputs (tuple of Variable): inputs to the function
+        grad_outputs (tuple of Variable, optional): The gradients with respect to
+            the function's outputs.
+        eps (float, optional): perturbation for finite differences
+        atol (float, optional): absolute tolerance
+        rtol (float, optional): relative tolerance
 
     Returns:
-        True if all differences satisfy allclose condition
+        True if all differences satisfy allclose condition. Raises an exception
+        otherwise.
     """
+    if grad_outputs is None:
+        # If grad_outputs is not specified, create random variables of the same
+        # shape, type, and device as the outputs
+        def randn_like(x):
+            return Variable(x.data.new(x.size()).normal_(), requires_grad=True)
+        outputs = _as_tuple(func(*inputs))
+        grad_outputs = [randn_like(x) for x in outputs]
+
     def new_func(*input_args):
         input_args = input_args[:-len(grad_outputs)]
         outputs = _differentiable_outputs(func(*input_args))

torch/autograd/profiler.py

@@ -17,6 +17,17 @@ class EventList(list):
         return self.table()
 
     def table(self, sort_by=None):
+        """Prints an EventList as a nicely formatted table.
+
+        Arguments:
+            sort_by (str, optional): Attribute used to sort entries. By default
+                they are printed in the same order as they were registered.
+                Valid keys include: ``cpu_time``, ``cuda_time``, ``cpu_time_total``,
+                ``cuda_time_total``, ``count``.
+
+        Returns:
+            A string containing the table.
+        """
         return build_table(self, sort_by)
 
     def export_chrome_trace(self, path):
@@ -72,7 +83,7 @@ class profile(object):
 
     Arguments:
         enabled (bool, optional): Setting this to False makes this context manager a no-op.
-            Default: True.
+            Default: ``True``.
 
     .. warning:
         This context managers should not be called recursively, i.e. at most one
@@ -131,6 +142,12 @@ class profile(object):
             return '<unfinished torch.autograd.profile>'
         return str(self.function_events)
 
+    def table(self, sort_by=None):
+        if self.function_events is None:
+            raise RuntimeError("can't export a trace that didn't finish running")
+        return self.function_events.table(sort_by)
+    table.__doc__ = EventList.table.__doc__
+
     def export_chrome_trace(self, path):
         if self.function_events is None:
             raise RuntimeError("can't export a trace that didn't finish running")
@@ -153,18 +170,24 @@ class profile(object):
 class emit_nvtx(object):
     """Context manager that makes every autograd operation emit an NVTX range.
 
-    It is useful when running the program under nvprof. Unfortunately, there's no
-    way to force nvprof to flush the data it collected to disk, so for CUDA profiling
-    one has to use this context manager to annotate nvprof traces, and then use
-    :func:`torch.autograd.profiler.open_nvtx` to analyze the checkpoint.
+    It is useful when running the program under nvprof::
+
+        nvprof --profile-from-start off -o trace_name.prof -- <regular command here>
+
+    Unfortunately, there's no way to force nvprof to flush the data it collected
+    to disk, so for CUDA profiling one has to use this context manager to annotate
+    nvprof traces and wait for the process to exit before inspecting them.
+    Then, either NVIDIA Visual Profiler (nvvp) can be used to visualize the timeline, or
+    :func:`torch.autograd.profiler.load_nvprof` can load the results for inspection
+    e.g. in Python REPL.
 
     .. warning:
-        This context managers should not be called recursively, i.e. at most one
+        This context manager should not be called recursively, i.e. at most one
         instance should be enabled at any given time.
 
     Arguments:
         enabled (bool, optional): Setting this to False makes this context manager a no-op.
-            Default: True.
+            Default: ``True``.
 
     Example:
         >>> with torch.cuda.profiler.profile():
@@ -291,7 +314,7 @@ def demangle(name):
     try:
         with open(os.devnull, 'w') as devnull:
             return subprocess.check_output(['c++filt', '-n', name], stderr=devnull).rstrip().decode("ascii")
-    except subprocess.CalledProcessError:
+    except (subprocess.CalledProcessError, OSError, FileNotFoundError) as e:
         return name
 
 

torch/autograd/variable.py

@@ -154,14 +154,14 @@ class Variable(_C._VariableBase):
                 None values can be specified for scalar Variables or ones that
                 don't require grad. If a None value would be acceptable then
                 this argument is optional.
-            retain_graph (bool, optional): If False, the graph used to compute
+            retain_graph (bool, optional): If ``False``, the graph used to compute
                 the grads will be freed. Note that in nearly all cases setting
                 this option to True is not needed and often can be worked around
                 in a much more efficient way. Defaults to the value of
                 ``create_graph``.
-            create_graph (bool, optional): If true, graph of the derivative will
+            create_graph (bool, optional): If ``True``, graph of the derivative will
                 be constructed, allowing to compute higher order derivative
-                products. Defaults to False, unless ``gradient`` is a volatile
+                products. Defaults to ``False``, unless ``gradient`` is a volatile
                 Variable.
         """
         torch.autograd.backward(self, gradient, retain_graph, create_graph, retain_variables)
@@ -205,20 +205,31 @@ class Variable(_C._VariableBase):
         return handle
 
     def reinforce(self, reward):
-        """Registers a reward obtained as a result of a stochastic process.
+        def trim(str):
+            return '\n'.join([line.strip() for line in str.split('\n')])
 
-        Differentiating stochastic nodes requires providing them with reward
-        value. If your graph contains any stochastic operations, you should
-        call this function on their outputs. Otherwise an error will be raised.
+        raise RuntimeError(trim(r"""reinforce() was removed.
+            Use torch.distributions instead.
+            See http://pytorch.org/docs/master/distributions.html
 
-        Parameters:
-            reward(Tensor): Tensor with per-element rewards. It has to match
-                the device location and shape of Variable's data.
-        """
-        if not isinstance(self.grad_fn, StochasticFunction):
-            raise RuntimeError("reinforce() can be only called on outputs "
-                               "of stochastic functions")
-        self.grad_fn._reinforce(reward)
+            Instead of:
+
+            probs = policy_network(state)
+            action = probs.multinomial()
+            next_state, reward = env.step(action)
+            action.reinforce(reward)
+            action.backward()
+
+            Use:
+
+            probs = policy_network(state)
+            # NOTE: categorical is equivalent to what used to be called multinomial
+            m = torch.distributions.Categorical(probs)
+            action = m.sample()
+            next_state, reward = env.step(action)
+            loss = -m.log_prob(action) * reward
+            loss.backward()
+        """))
 
     def detach(self):
         """Returns a new Variable, detached from the current graph.
@@ -422,7 +433,7 @@ class Variable(_C._VariableBase):
         return self.expand(tensor.size())
 
     def multinomial(self, num_samples=1, replacement=False):
-        return Multinomial.apply(self, num_samples, replacement)
+        return Categorical.apply(self, num_samples, replacement)
 
     def bernoulli(self):
         return Bernoulli.apply(self)

torch/backends/cudnn/__init__.py

@@ -257,10 +257,11 @@ class RNNDescriptor(object):
                 CUDNN_RNN_ALGO_STANDARD,
                 datatype
             ))
-        if version() >= 7000 and int(cuda[0]) >= 9:
-            lib.cudnnSetRNNMatrixMathType(self, CUDNN_DEFAULT_MATH)
-            if datatype == CUDNN_DATA_HALF:
-                lib.cudnnSetRNNMatrixMathType(self, CUDNN_TENSOR_OP_MATH)
+            if version() >= 7000 and int(cuda[0]) >= 9 and (
+                    torch.cuda.get_device_capability(torch.cuda.current_device())[0] >= 7):
+                lib.cudnnSetRNNMatrixMathType(self, CUDNN_DEFAULT_MATH)
+                if datatype == CUDNN_DATA_HALF:
+                    lib.cudnnSetRNNMatrixMathType(self, CUDNN_TENSOR_OP_MATH)
         else:
             check_error(lib.cudnnSetRNNDescriptor(
                 self,

torch/csrc/Exceptions.cpp

@@ -1,5 +1,8 @@
 #include <Python.h>
 
+#include <utility>
+#include <vector>
+
 #include "THP.h"
 
 PyObject *THPException_FatalError;
@@ -11,3 +14,61 @@ bool THPException_init(PyObject *module)
   ASSERT_TRUE(PyModule_AddObject(module, "FatalError", THPException_FatalError) == 0);
   return true;
 }
+
+namespace torch {
+
+void replaceAll(std::string & str,
+    const std::string & old_str,
+    const std::string & new_str) {
+  std::string::size_type pos = 0u;
+  while ((pos = str.find(old_str, pos)) != std::string::npos){
+     str.replace(pos, old_str.length(), new_str);
+  }
+}
+
+std::string processErrorMsg(std::string str) {
+
+  // Translate Aten types to their respective pytorch ones
+  std::vector<std::pair<std::string, std::string>> changes {
+    {"SparseCUDAByteType", "torch.cuda.sparse.ByteTensor"},
+    {"SparseCUDACharType", "torch.cuda.sparse.CharTensor"},
+    {"SparseCUDADoubleType", "torch.cuda.sparse.DoubleTensor"},
+    {"SparseCUDAFloatType", "torch.cuda.sparse.FloatTensor"},
+    {"SparseCUDAIntType", "torch.cuda.sparse.IntTensor"},
+    {"SparseCUDALongType", "torch.cuda.sparse.LongTensor"},
+    {"SparseCUDAShortType", "torch.cuda.sparse.ShortTensor"},
+    {"SparseCUDAHalfType", "torch.cuda.sparse.HalfTensor"},
+    {"SparseCPUByteType", "torch.sparse.ByteTensor"},
+    {"SparseCPUCharType", "torch.sparse.CharTensor"},
+    {"SparseCPUDoubleType", "torch.sparse.DoubleTensor"},
+    {"SparseCPUFloatType", "torch.sparse.FloatTensor"},
+    {"SparseCPUIntType", "torch.sparse.IntTensor"},
+    {"SparseCPULongType", "torch.sparse.LongTensor"},
+    {"SparseCPUShortType", "torch.sparse.ShortTensor"},
+    {"SparseCPUHalfType", "torch.sparse.HalfTensor"},
+    {"CUDAByteType", "torch.cuda.ByteTensor"},
+    {"CUDACharType", "torch.cuda.CharTensor"},
+    {"CUDADoubleType", "torch.cuda.DoubleTensor"},
+    {"CUDAFloatType", "torch.cuda.FloatTensor"},
+    {"CUDAIntType", "torch.cuda.IntTensor"},
+    {"CUDALongType", "torch.cuda.LongTensor"},
+    {"CUDAShortType", "torch.cuda.ShortTensor"},
+    {"CUDAHalfType", "torch.cuda.HalfTensor"},
+    {"CPUByteType", "torch.ByteTensor"},
+    {"CPUCharType", "torch.CharTensor"},
+    {"CPUDoubleType", "torch.DoubleTensor"},
+    {"CPUFloatType", "torch.FloatTensor"},
+    {"CPUIntType", "torch.IntTensor"},
+    {"CPULongType", "torch.LongTensor"},
+    {"CPUShortType", "torch.ShortTensor"},
+    {"CPUHalfType", "torch.HalfTensor"},
+  };
+
+  for (const auto & it : changes) {
+    replaceAll(str, it.first, it.second);
+  }
+
+  return str;
+}
+}
+

torch/csrc/Exceptions.h

@@ -14,7 +14,8 @@
   } catch (python_error &e) {                                                  \
     return retval;                                                             \
   } catch (std::exception &e) {                                                \
-    PyErr_SetString(PyExc_RuntimeError, e.what());                             \
+    auto msg = torch::processErrorMsg(e.what());                               \
+    PyErr_SetString(PyExc_RuntimeError, msg.c_str());                          \
     return retval;                                                             \
   }
 
@@ -68,4 +69,8 @@ struct python_error : public std::exception {
 bool THPException_init(PyObject *module);
 #endif
 
+namespace torch {
+std::string processErrorMsg(std::string str);
+}
+
 #endif

torch/csrc/Storage.cpp

@@ -1,3 +1,5 @@
+#define __STDC_FORMAT_MACROS
+
 #include <Python.h>
 #include <structmember.h>
 

torch/csrc/Tensor.cpp

@@ -1,3 +1,5 @@
+#define __STDC_FORMAT_MACROS
+
 #include <Python.h>
 #include <structmember.h>
 

torch/csrc/autograd/functions/batch_normalization.cpp

@@ -164,7 +164,7 @@ auto BatchNormBackward::apply(const variable_list& grad_outputs) -> variable_lis
   // Add saved variables used out of the pure autograd to inputs
   variable_list all_inputs(grad_outputs);
   all_inputs.push_back(input_var);
-  if (weight.get()) {
+  if (weight.defined()) {
     all_inputs.push_back(weight_var);
   }
   auto outputs =  as_tensor_list(std::move(grad_input),

torch/csrc/autograd/functions/convolution.cpp

@@ -365,7 +365,7 @@ auto ConvForward::apply(const variable_list& inputs) -> variable_list {
 // For Convolution strategies that don't implicitly handle grad_bias, we add a helper
 // function here to perform it using simple Tensor operators
 static at::Tensor compute_grad_bias(const at::Tensor& grad_output) {
-  // grad_output is in N, C, H, W, we re-shape and reduce over spatial dims and batches 
+  // grad_output is in N, C, H, W, we re-shape and reduce over spatial dims and batches
   return grad_output.contiguous().view({grad_output.size(0), grad_output.size(1), -1}).sum(0).sum(1);
 }
 
@@ -727,7 +727,18 @@ auto ConvBackwardBackward::apply(const variable_list& grad_grad_inputs) -> varia
       gI = apply_fn<Transpose>(0, 1)(gIt);
     }
   }
-  return {ggO, gI, gW};
+
+  if (should_compute_output(0) && !ggO.defined()) ggO = at::zeros_like(gO);
+  if (should_compute_output(1) && !gI.defined()) gI = at::zeros_like(input);
+  if (should_compute_output(2) && !gW.defined()) gW = at::zeros_like(weight);
+  bool is_volatile = std::any_of(grad_grad_inputs.begin(), grad_grad_inputs.end(), [](const Variable& v){
+    return v.defined() && v.is_volatile();
+  });
+  auto results = variable_list({ggO, gI, gW});
+  for (auto& result : results) {
+    result.is_volatile() |= is_volatile;
+  }
+  return results;
 }
 
 auto ConvBackwardBackward::releaseVariables() -> void {

torch/csrc/autograd/functions/onnx/batch_normalization.cpp

@@ -9,7 +9,7 @@ namespace autograd {
 jit::node_list BatchNormForward::symbolic(SymbolicContext* ctx, jit::node_list inputs) {
   auto & g = ctx->graph;
   // X, Scale, Bias
-  auto bn = g->appendNode(g->create(jit::kSpatialBN,{inputs.at(0),inputs.at(1),inputs.at(2)}));
+  auto bn = g->appendNode(g->create(jit::kBatchNormalization, {inputs.at(0),inputs.at(1),inputs.at(2)}));
   bn->addInput(jit::tracer::getBufferTrace(*ctx->buffer_map, running_mean));
   bn->addInput(jit::tracer::getBufferTrace(*ctx->buffer_map, running_var));
   bn->i_(jit::kis_test, !this->training);

torch/csrc/autograd/functions/onnx/convolution.cpp

@@ -18,7 +18,7 @@ namespace torch { namespace autograd {
 jit::node_list ConvForward::symbolic(SymbolicContext* ctx, jit::node_list inputs) {
   auto & g = ctx->graph;
   // See Note [Caffe2ConvTranspose]
-  auto n = g->create(!transposed ? jit::kConv : jit::kCaffe2ConvTranspose,
+  auto n = g->create(!transposed ? jit::kConv : jit::kConvTranspose,
                                    {inputs.at(0), inputs.at(1)});
 
   // Irritatingly, Caffe2 requires us to specify kernels,
@@ -55,6 +55,8 @@ jit::node_list ConvForward::symbolic(SymbolicContext* ctx, jit::node_list inputs
   n->i_(jit::kgroup,groups);
 
   // Not in ONNX?
+  // TODO: implement it once ConvTranspose in ONNX gets `adj` argument instead
+  // of providing `output_shape`
   for (int p : output_padding) {
     JIT_EXPECTM(p == 0, "output padding is not supported.");
   }

torch/csrc/autograd/input_buffer.cpp

@@ -1,5 +1,6 @@
 #include "torch/csrc/autograd/input_buffer.h"
 
+#include "torch/csrc/assertions.h"
 #include "torch/csrc/autograd/functions/basic_ops.h"
 #include "torch/csrc/utils/auto_gpu.h"
 
@@ -10,6 +11,7 @@ InputBuffer::InputBuffer(size_t size)
   {}
 
 void InputBuffer::add(size_t pos, Variable var) {
+  TORCH_ASSERT(pos >= 0 && pos < buffer.size());
   if (!var.defined()) {
     return;
   }

torch/csrc/autograd/python_variable.cpp

@@ -43,6 +43,10 @@ PyObject * THPVariable_Wrap(Variable var)
     Py_RETURN_NONE;
   }
 
+  if (var.dim() == 0) {
+    throw std::runtime_error("Variable API does not support Scalars");
+  }
+
   if (auto obj = var.get()->pyobj) {
     Py_INCREF(obj);
     return obj;

torch/csrc/autograd/utils/wrap_outputs.h

@@ -13,6 +13,10 @@
 namespace torch { namespace autograd { namespace utils {
 
 inline PyObject* wrap(at::Tensor tensor) {
+  if (tensor.defined() && tensor.dim() == 0) {
+    // don't expose 0-dim tensors to Variable API.
+    Variable(tensor).data().as_strided_({1}, {1});
+  }
   return THPVariable_Wrap(Variable(std::move(tensor)));
 }
 
@@ -54,6 +58,10 @@ inline PyObject* wrap(int64_t value) {
   return THPUtils_packInt64(value);
 }
 
+inline PyObject* wrap(void* value) {
+  return THPUtils_packInt64(reinterpret_cast<intptr_t>(value));
+}
+
 inline PyObject* wrap(at::Scalar scalar) {
   return wrap(scalar.toTensor());
 }

torch/csrc/cuda/Module.cpp

@@ -133,6 +133,18 @@ PyObject * THCPModule_getDeviceName_wrap(PyObject *self, PyObject *arg)
   END_HANDLE_TH_ERRORS
 }
 
+PyObject * THCPModule_getDeviceCapability_wrap(PyObject *self, PyObject *arg)
+{
+  HANDLE_TH_ERRORS
+  THPUtils_assert(THPUtils_checkLong(arg), "invalid argument to getDeviceCapability");
+  long device = THPUtils_unpackLong(arg);
+
+  cudaDeviceProp prop;
+  THCudaCheck(cudaGetDeviceProperties(&prop, device));
+  return Py_BuildValue("(ii)", prop.major, prop.minor);
+  END_HANDLE_TH_ERRORS
+}
+
 PyObject * THCPModule_getCurrentStream_wrap(PyObject *self)
 {
   HANDLE_TH_ERRORS
@@ -174,6 +186,11 @@ PyObject * THCPModule_getDriverVersion(PyObject *self)
   return PyLong_FromLong((long) driverVersion);
 }
 
+PyObject * THCPModule_getCompiledVersion(PyObject *self)
+{
+  return PyLong_FromLong((long) CUDA_VERSION);
+}
+
 PyObject * THCPModule_getRNGState(PyObject *_unused)
 {
   HANDLE_TH_ERRORS
@@ -297,6 +314,15 @@ PyObject * THCPModule_cudaUnlockMutex(PyObject *module)
   Py_RETURN_NONE;
 }
 
+PyObject * THCPModule_emptyCache(PyObject *_unused)
+{
+  HANDLE_TH_ERRORS
+  auto device_allocator = THCState_getDeviceAllocator(state);
+  THCudaCheck(device_allocator->emptyCache(device_allocator->state));
+  END_HANDLE_TH_ERRORS
+  Py_RETURN_NONE;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // Cuda module initialization
 ////////////////////////////////////////////////////////////////////////////////
@@ -369,13 +395,16 @@ static struct PyMethodDef _THCPModule_methods[] = {
   {"_cuda_getDevice",   (PyCFunction)THCPModule_getDevice_wrap,   METH_NOARGS,  NULL},
   {"_cuda_getDeviceCount", (PyCFunction)THCPModule_getDeviceCount_wrap, METH_NOARGS, NULL},
   {"_cuda_getDeviceName", (PyCFunction)THCPModule_getDeviceName_wrap, METH_O,   NULL},
+  {"_cuda_getDeviceCapability", (PyCFunction)THCPModule_getDeviceCapability_wrap, METH_O,   NULL},
   {"_cuda_getCurrentStream", (PyCFunction)THCPModule_getCurrentStream_wrap, METH_NOARGS, NULL},
   {"_cuda_getCurrentBlasHandle", (PyCFunction)THCPModule_getCurrentBlasHandle_wrap, METH_NOARGS, NULL},
   {"_cuda_setStream",    (PyCFunction)THCPModule_setStream_wrap,  METH_O, NULL},
   {"_cuda_isDriverSufficient", (PyCFunction)THCPModule_isDriverSufficient, METH_NOARGS, NULL},
   {"_cuda_getDriverVersion", (PyCFunction)THCPModule_getDriverVersion, METH_NOARGS, NULL},
+  {"_cuda_getCompiledVersion", (PyCFunction)THCPModule_getCompiledVersion, METH_NOARGS, NULL},
   {"_cuda_getRNGState", (PyCFunction)THCPModule_getRNGState,      METH_NOARGS,  NULL},
   {"_cuda_setRNGState", (PyCFunction)THCPModule_setRNGState,      METH_O,       NULL},
+  {"_cuda_emptyCache", (PyCFunction) THCPModule_emptyCache,       METH_NOARGS,  NULL},
   {"_cuda_manualSeed",  (PyCFunction)THCPModule_manualSeed,       METH_O,       NULL},
   {"_cuda_manualSeedAll", (PyCFunction)THCPModule_manualSeedAll,  METH_O,       NULL},
   {"_cuda_seed",        (PyCFunction)THCPModule_seed,             METH_NOARGS,  NULL},

torch/csrc/cuda/Storage.cpp

@@ -1,3 +1,5 @@
+#define __STDC_FORMAT_MACROS
+
 #include <Python.h>
 #include <structmember.h>
 

torch/csrc/cuda/Tensor.cpp

@@ -1,3 +1,5 @@
+#define __STDC_FORMAT_MACROS
+
 #include <Python.h>
 #include <structmember.h>
 

torch/csrc/cudnn/Conv.cpp

@@ -228,12 +228,12 @@ struct algorithm_search<cudnnConvolutionFwdAlgo_t> {
         conv.cdesc.desc,
         conv.odesc.desc,
         out,
-        1,
+        n_algo,
         &algoCount,
         perfResults,
         ws.data,
         ws.size));
-    return getBestAlgorithm<cudnnConvolutionFwdAlgoPerf_t>(perfResults, deterministic, n_algo);
+    return getBestAlgorithm<cudnnConvolutionFwdAlgoPerf_t>(perfResults, deterministic, algoCount);
   }
 
   static void getAlgorithm(
@@ -302,12 +302,12 @@ struct algorithm_search<cudnnConvolutionBwdDataAlgo_t> {
         conv.cdesc.desc,
         conv.idesc.desc,
         in,
-        1,
+        n_algo,
         &algoCount,
         perfResults,
         ws.data,
         ws.size));
-    return getBestAlgorithm<cudnnConvolutionBwdDataAlgoPerf_t>(perfResults, deterministic, n_algo);
+    return getBestAlgorithm<cudnnConvolutionBwdDataAlgoPerf_t>(perfResults, deterministic, algoCount);
   }
 
   static void getAlgorithm(cudnnHandle_t handle, const Convolution& conv, cudnnConvolutionBwdDataAlgo_t* algo) {
@@ -376,12 +376,12 @@ struct algorithm_search<cudnnConvolutionBwdFilterAlgo_t> {
         conv.cdesc.desc,
         conv.wdesc.desc,
         wght,
-        1,
+        n_algo,
         &algoCount,
         perfResults,
         ws.data,
         ws.size));
-    return getBestAlgorithm<cudnnConvolutionBwdFilterAlgoPerf_t>(perfResults, deterministic, n_algo);
+    return getBestAlgorithm<cudnnConvolutionBwdFilterAlgoPerf_t>(perfResults, deterministic, algoCount);
   }
 
   static void getAlgorithm(

torch/csrc/generic/methods/Tensor.cwrap

@@ -761,6 +761,12 @@ PyObject * THPTensor_(stride)(PyObject *self, PyObject *args, PyObject *kwargs)
         - accreal start
         - accreal end
         - CONSTANT 1
+      - arguments:
+        - arg: THTensor* result
+          output: True
+        - CONSTANT 0
+        - accreal end
+        - CONSTANT 1
 ]]
 
 [[

torch/csrc/jit/attributes.h

@@ -78,7 +78,7 @@ using GraphsAttr = VectorAttributeValue<std::shared_ptr<Graph>,AttributeKind::gs
 
 // CRTP so that Node which inherits Attributes can be return for
 // method chaining e.g:
-// Node * n = g->create(kSelect)->set_i(kOffset,3)->set_f(kValue,3.5);
+// Node * n = g->create(kSelect)->i_(kOffset,3)->f_(kValue,3.5);
 // we return Derived* pointers because Nodes are normally held as pointers.
 template<typename Derived>
 struct Attributes {

torch/csrc/jit/export.cpp

@@ -69,7 +69,8 @@ void encodeTensor(onnx::TensorProto * p, const at::Tensor & tensor) {
       break;
   }
   p->set_data_type(onnx_type);
-  at::Tensor cont = tensor.toType(at::CPU(at_type)).contiguous();
+  // CPU's HalfTensor doesn't have contiguous(), so first calling contiguous()
+  at::Tensor cont = tensor.contiguous().toType(at::CPU(at_type));
   p->set_raw_data(cont);
 }
 
@@ -79,40 +80,50 @@ void addAttribute(onnx::NodeProto * n_p, jit::Node * n, jit::Symbol name) {
   switch(n->kindOf(name)) {
     case AttributeKind::f:
       attr->set_f(n->f(name));
+      attr->set_type(onnx::aFLOAT);
       break;
     case AttributeKind::fs:
+      attr->set_type(onnx::aFLOATS);
       for(auto & v : n->fs(name))
         attr->add_floats(v);
       break;
     case AttributeKind::i:
+      attr->set_type(onnx::aINT);
       attr->set_i(n->i(name));
       break;
     case AttributeKind::is:
+      attr->set_type(onnx::aINTS);
       for(auto & v : n->is(name))
         attr->add_ints(v);
       break;
     case AttributeKind::s:
+      attr->set_type(onnx::aSTRING);
       attr->set_s(n->s(name));
       break;
     case AttributeKind::ss:
+      attr->set_type(onnx::aSTRINGS);
       for(auto & v : n->ss(name))
         attr->add_strings(v);
       break;
     case AttributeKind::t: {
+      attr->set_type(onnx::aTENSOR);
       auto t = attr->mutable_t();
       encodeTensor(t, n->t(name));
     } break;
     case AttributeKind::ts:
+      attr->set_type(onnx::aTENSORS);
       for(auto & v : n->ts(name)) {
         auto t = attr->add_tensors();
         encodeTensor(t, v);
       }
       break;
     case AttributeKind::g: {
+      attr->set_type(onnx::aGRAPH);
       auto g = attr->mutable_g();
       encodeGraph(g, n->g(name), {});
     } break;
     case AttributeKind::gs:
+      attr->set_type(onnx::aGRAPHS);
       for(auto & v : n->gs(name)) {
         auto g = attr->add_graphs();
         encodeGraph(g, v, {});
@@ -191,6 +202,9 @@ void encodeGraph(onnx::GraphProto * p_g, const std::shared_ptr<Graph> & g, const
       continue;
     }
     auto p_n = p_g->add_node();
+    if (node->getSourceLocation()) {
+      p_n->set_doc_string(node->getSourceLocation()->python_traceback);
+    }
     for(auto input : node->inputs()) {
       p_n->add_input(node_name(input));
     }
@@ -256,11 +270,18 @@ void validateGraph(const std::shared_ptr<Graph>& graph) {
 }
 
 std::string ExportGraph(const std::shared_ptr<Graph>& graph,
-                        const std::vector<at::Tensor> & initializers) {
+                        const std::vector<at::Tensor> & initializers,
+                        int64_t onnx_opset_version) {
 
   validateGraph(graph);
 
   onnx::ModelProto model_proto;
+  model_proto.set_producer_name("pytorch");
+  model_proto.set_producer_version("0.3");
+  auto* imp = model_proto.add_opset_import();
+  // This is the version of ONNX operator set we are targeting
+  imp->set_version(onnx_opset_version);
+
   // Set up nanopb callbacks and compute the amount of space needed to store
   // the resulting protobuf
   encodeModel(&model_proto, graph, initializers);

torch/csrc/jit/export.h

@@ -5,6 +5,7 @@
 namespace torch { namespace jit {
 
 std::string ExportGraph(const std::shared_ptr<Graph>& graph,
-                        const std::vector<at::Tensor> & initializers);
+                        const std::vector<at::Tensor> & initializers,
+                        int64_t onnx_opset_version);
 
 }}

torch/csrc/jit/fusion_compiler.cpp

@@ -261,6 +261,14 @@ CompiledFusionFunction::CompiledFusionFunction(const std::string & name, Annotat
   , output_desc(agraph.output_desc) {
   JIT_CUDA_CHECK(cudaGetDevice(&device));
   JIT_CUDA_CHECK(cudaGetDeviceProperties(&prop, device));
+  if ((prop.major >= 6 && CUDA_VERSION < 8000) ||
+      (prop.major >= 7 && CUDA_VERSION < 9000)) {
+    std::stringstream err_string;
+    err_string << "PyTorch compiled with insufficient CUDA version: " 
+	       << CUDA_VERSION << " for the current GPU device " << prop.name 
+	       << " with device capability " << prop.major << "." << prop.minor;
+    throw std::runtime_error(err_string.str());
+  }
 
   std::stringstream cu;
   concat_desc = codegen::emitCompilationUnit(cu, name, agraph);

torch/csrc/jit/interned_strings.h

@@ -43,9 +43,8 @@ _(split) \
 _(Offset) \
 _(value) \
 _(Subgraph) \
-_(SpatialBN) \
+_(BatchNormalization) \
 _(Conv) \
-_(Caffe2ConvTranspose) \
 _(ConvTranspose) \
 _(is_test) \
 _(epsilon) \
@@ -75,6 +74,8 @@ _(shape) \
 _(axes) \
 _(group) \
 _(inplace) \
+_(transA) \
+_(transB) \
 _(other)
 
 enum BuiltinSymbol {

torch/csrc/jit/ir.cpp

@@ -41,6 +41,7 @@ void printNodeRef(std::ostream & out, const Node * n) {
 template <typename T>
 std::ostream& operator<<(std::ostream & out, const std::vector<T> & nodes) {
   out << at::ArrayRef<T>{nodes};
+  return out;
 }
 
 template <typename T>

torch/csrc/jit/ir.h

@@ -60,6 +60,15 @@ static inline bool operator==(const Use & a, const Use & b) {
 // Graph holds a list of parameters.
 struct Param;
 
+// SourceLocation represents source code-level debug information for a node.
+// It contains a Python stack trace that represents the provenance of a given
+// node in the trace.
+struct SourceLocation {
+  SourceLocation(std::string python_traceback)
+  : python_traceback(std::move(python_traceback)) {}
+  std::string python_traceback;
+};
+
 // the list types are intentionally simple, but we type-def
 // them here so if we need to change them, refactoring will be easier
 using node_list = std::vector<Node*>;
@@ -113,6 +122,7 @@ private:
   size_t unique_ = 0;          // unique id
   size_t stage_ = 0;           // 0-forward, 1-backward, 2-double-backward,...
   std::string debug_name_;
+  std::shared_ptr<SourceLocation> source_location_;
 protected:
   TypePtr type_;
   Node(Graph * graph_, NodeKind kind_); //defined after graph
@@ -150,6 +160,13 @@ public:
   const std::string & debugName() const {
     return debug_name_;
   }
+  Node* setSourceLocation(std::shared_ptr<SourceLocation> sl) {
+    source_location_ = sl;
+    return this;
+  }
+  std::shared_ptr<SourceLocation> getSourceLocation() const {
+    return source_location_;
+  }
   Graph * owningGraph() {
     return graph_;
   }
@@ -514,6 +531,7 @@ protected:
   virtual void cloneFrom(Node * s) {
     if (s->hasType()) setType(s->type());
     setDebugName(s->debugName());
+    setSourceLocation(s->getSourceLocation());
     copyAttributes(*s);
   }
 };

torch/csrc/jit/passes/onnx.cpp

@@ -86,6 +86,9 @@ void ToONNX(std::shared_ptr<tracer::TracingState>& state) {
         if (!outputs[i]->hasType()) {
           outputs[i]->setType(old->typeOption());
         }
+        // Copy over source location information to all nodes created by
+        // the symbolic
+        outputs[i]->setSourceLocation(node->getSourceLocation());
         env[old] = outputs[i];
       } else {
         // Null output means that the ONNX op doesn't have outputs corresponding
@@ -121,6 +124,31 @@ void ToONNX(std::shared_ptr<tracer::TracingState>& state) {
     }
   };
 
+  // Cast output of symbolic() python implementation
+  auto processSymbolicOutput = [&](const std::string& op_name, Node* n, const py::object& raw_output) {
+    if (raw_output.ptr() == Py_None) {
+      cloneNode(n);
+      return;
+    }
+    // Cast the outputs back to C++ and put them in the new graph
+    std::vector<Node*> outputs;
+    try {
+      if (py::isinstance<Node>(raw_output)) {
+        outputs = node_list{py::cast<Node*>(raw_output)};
+      } else {
+        outputs = py::cast<std::vector<Node*>>(raw_output);
+      }
+    } catch (const std::exception& ex) {
+      std::ostringstream ss;
+      ss << "Error casting results of symbolic for " << op_name
+         << ": expected to return list of op nodes, instead received type ''"
+         << py::str(raw_output.get_type()) << "': " << py::str(raw_output);
+      throw std::runtime_error(ss.str());
+    }
+
+    setOutputs(op_name, n, outputs);
+  };
+
   auto callPySymbolicFunction = [&](Node* n) {
     // The idea is delegate as much of the actual argument massaging to
     // Python as possible
@@ -133,19 +161,7 @@ void ToONNX(std::shared_ptr<tracer::TracingState>& state) {
 
     py::object raw_output = onnx.attr("_run_symbolic_function")(ctx.graph, n, py_inputs);
 
-    if (raw_output.ptr() == Py_None) {
-      cloneNode(n);
-    } else {
-      // Cast the outputs back to C++ and put them in the new graph
-      node_list outputs;
-      if (py::isinstance<Node>(raw_output)) {
-        outputs = node_list{py::cast<Node*>(raw_output)};
-      } else {
-        outputs = py::cast<std::vector<Node*>>(raw_output);
-      }
-
-      setOutputs(symbolToString(n->kind()), n, outputs);
-    }
+    processSymbolicOutput(symbolToString(n->kind()), n, raw_output);
   };
 
   auto callPySymbolicMethod = [&](PythonOp* op) {
@@ -184,20 +200,7 @@ void ToONNX(std::shared_ptr<tracer::TracingState>& state) {
     // upon argument mismatch
     py::object raw_output = onnx.attr("_run_symbolic_method")(op->name(), pyobj.attr("symbolic"), py_symbolic_args);
 
-    if (raw_output.ptr() == Py_None) {
-      cloneNode(op);
-      return;
-    }
-
-    // Cast the outputs back to C++ and put them in the new graph
-    std::vector<Node*> outputs;
-    if (py::isinstance<Node>(raw_output)) {
-      outputs = node_list{py::cast<Node*>(raw_output)};
-    } else {
-      outputs = py::cast<std::vector<Node*>>(raw_output);
-    }
-
-    setOutputs(op->name(), op, outputs);
+    processSymbolicOutput(op->name(), op, raw_output);
   };
 
   // Finally, visit all nodes in the graph

torch/csrc/jit/passes/onnx/peephole.cpp

@@ -15,24 +15,62 @@ std::unordered_set<NodeKind> broadcasting = {
   kGemm,
 };
 
+bool isNopTranspose(const std::vector<int64_t> & perm) {
+  for (size_t i = 0; i < perm.size(); i++)
+    if (perm[i] != i)
+      return false;
+  return true;
+}
+
+// returns a vector `ret` such that transposing by `ret` is equivalent
+// to transposing by `t1` and then by `t2`
+std::vector<int64_t> composeTransposes(const std::vector<int64_t> & t1,
+                                       const std::vector<int64_t> & t2) {
+  JIT_ASSERT(t1.size() == t2.size());
+  std::vector<int64_t> ret;
+  for (size_t i = 0; i < t1.size(); i++) {
+    JIT_ASSERT(   t1[i]  < t2.size());
+    JIT_ASSERT(t2[t1[i]] < t2.size());
+    ret.push_back(t2[t1[i]]);
+  }
+  return ret;
+}
+
 bool isBroadcasting(Node *node) {
   return broadcasting.count(node->kind());
 }
 
-// When iterating over the dimension sizes, starting at the trailing dimension,
-// the dimension sizes must either be equal, or one of them does not exist.
+// First iterate over the 'from' tensor sizes. Ignore all leading and trailing
+// dimensions that are simply one, since they can be trivially broadcasted.
+// When iterating over the dimension sizes (with reduced 'from' tensor),
+// starting at the trailing dimension, the dimension sizes must either be equal,
+// or one of them does not exist.
 //
-//  equivalently:
-//
-// Test that 'from' is a suffix of 'to'.
+// Note that this is NOT equivalent to numpy broadcasting semantics, and do
+// not represent that generalized broadcasting that Pytorch implements in
+// general. Rather, this is Caffe2-style broadcasting.
 bool fusibleExpandTo(at::IntList from, at::IntList to) {
-  auto f = from.rbegin();
-  auto t = to.rbegin();
-  for (; f != from.rend() && t != to.rend(); f++, t++) {
-    // TODO: if 1->n expansion is supported, adjust this conditional.
-    if (*f != *t) return false;
+  if (from.size() > to.size()) {
+    return false;
   }
-  return f == from.rend();
+  ssize_t from_dim_start = 0, from_dim_end = from.size() - 1;
+  while (from_dim_start < from.size() && from[from_dim_start] == 1) {
+    from_dim_start++;
+  }
+  while (from_dim_end > from_dim_start && from[from_dim_end] == 1) {
+    from_dim_end--;
+  }
+
+  ssize_t f = from_dim_end;
+  ssize_t t = to.size() - 1;
+  for (; f >= from_dim_start && t >= 0; --f, --t) {
+    if (from[f] != to[t]) return false;
+  }
+
+  // In the case that the 'to' tensor has leading ones in the same place that
+  // the 'from' tensor does, f will be less than from_dim_start rather than
+  // strictly equal. E.x.: to := [5, 1, 768] and from := [1, 1, 768]
+  return f <= from_dim_start;
 }
 
 void fuseBroadcast(std::shared_ptr<Graph>& graph) {
@@ -76,6 +114,58 @@ void fuseBroadcast(std::shared_ptr<Graph>& graph) {
   }
 }
 
+void fuseConsecutiveTransposes(std::shared_ptr<Graph>& graph) {
+  for (auto it = graph->begin(); it != graph->end(); ++it) {
+    auto* n = *it;
+
+    if (n->kind() == kTranspose && n->input()->kind() == kTranspose) {
+      auto origInput = n->input();
+      n->is_(kperm, composeTransposes(origInput->is(kperm), n->is(kperm)));
+      n->replaceInput(0, origInput->input());
+      if (origInput->uses().size() == 0) {
+        origInput->destroy();
+      }
+      continue;
+    }
+  }
+}
+
+void eliminateNopTranspose(std::shared_ptr<Graph>& graph) {
+  for (auto it = graph->begin(); it != graph->end(); ++it) {
+    auto* n = *it;
+
+    if (n->kind() == kTranspose) {
+      if (isNopTranspose(n->is(kperm))) {
+        n->replaceAllUsesWith(n->input());
+        it.destroyCurrent();
+        continue;
+      }
+    }
+  }
+}
+
+void fuseTransposeIntoGemm(std::shared_ptr<Graph>& graph) {
+  static const std::vector<int64_t> simpleTransPerm({1,0});
+
+  for (auto it = graph->begin(); it != graph->end(); ++it) {
+    auto* n = *it;
+
+    if (n->kind() == kGemm) {
+      for (size_t i : {0,1}) {
+        auto inp = n->inputs()[i];
+        auto trans = i == 0 ? ktransA : ktransB;
+        if (inp->kind() == kTranspose && inp->is(kperm) == simpleTransPerm) {
+          n->replaceInput(i, inp->input());
+          n->i_(trans, n->hasAttribute(trans) ? !n->i(trans) : 1);
+          if (inp->uses().size() == 0) {
+            inp->destroy();
+          }
+        }
+      }
+    }
+  }
+}
+
 // This optimization does ONNX-specific peephole optimizations.
 //
 // At the moment, here are the optimizations it does:
@@ -83,6 +173,9 @@ void fuseBroadcast(std::shared_ptr<Graph>& graph) {
 //    easier for non-strided backends to more efficiently do broadcasts if this is
 //    local information.  This optimization is not useful for PyTorch as 'expand'
 //    is free.
+//  - Fusing of consecutive transposes
+//  - Elimiation of NOP transposes
+//  - Fusing of transposes into Gemm
 //
 // Before you write an optimization here, ask yourself, "Could I do this
 // optimization on ATen operators"?  If so, you should seriously consider
@@ -94,6 +187,9 @@ void PeepholeOptimizeONNX(std::shared_ptr<Graph>& graph) {
   // TODO: make it easier not to do O(k) iterations over the graph, where
   // k is the number of distinct peephole optimizations
   fuseBroadcast(graph);
+  fuseConsecutiveTransposes(graph);
+  eliminateNopTranspose(graph);
+  fuseTransposeIntoGemm(graph);
 }
 
 }}

torch/csrc/jit/passes/peephole.cpp

@@ -13,6 +13,7 @@ void PeepholeOptimize(std::shared_ptr<Graph>& graph) {
   for (auto it = graph->begin(); it != graph->end(); ++it) {
     auto* n = *it;
 
+    // eliminate redundant expand
     if (n->kind() == kexpand) {
       if (n->is(ksize) == n->input()->type()->expect<TensorType>()->sizes()) {
         n->replaceAllUsesWith(n->input());

torch/csrc/jit/python_tracer.cpp

@@ -32,13 +32,9 @@ void initPythonTracerBindings(PyObject* module_) {
       ss << *s.graph;
       return ss.str();
     })
-    .def("export", [](TracingState& s) {
+    .def("export", [](TracingState& s, const std::vector<at::Tensor>& initializers, int64_t onnx_opset_version) {
       ASSERT_UNEXPIRED("export");
-      return py::bytes(ExportGraph(s.graph, {}));
-    })
-    .def("export", [](TracingState& s, const std::vector<at::Tensor>& initializers) {
-      ASSERT_UNEXPIRED("export");
-      return py::bytes(ExportGraph(s.graph, initializers));
+      return py::bytes(ExportGraph(s.graph, initializers, onnx_opset_version));
     })
     .def("graph", [](TracingState& s) {
       return s.graph;

torch/csrc/jit/tracer.cpp

@@ -4,6 +4,11 @@
 #include "torch/csrc/autograd/function.h"
 #include "torch/csrc/autograd/python_engine.h"
 #include "torch/csrc/autograd/functions/special.h"
+#include "torch/csrc/utils/auto_gil.h"
+#include "torch/csrc/utils/python_strings.h"
+
+#include <frameobject.h>
+#include <patchlevel.h>
 
 namespace torch { namespace jit { namespace tracer {
 
@@ -89,6 +94,28 @@ void nontraceableBackwardSubgraph(const variable_list& inputs, const variable_li
   std::make_shared<autograd::Eval>()->replaceSubgraph(inputs, outputs);
 }
 
+namespace {
+// Python interpreter retrieval routine adapted from
+// https://stackoverflow.com/a/8706144
+std::string getPythonInterpreterStackTrace() {
+  std::stringstream stack_trace;
+  AutoGIL gil;
+  PyThreadState *tstate = PyThreadState_GET();
+  if (NULL != tstate && NULL != tstate->frame) {
+    PyFrameObject *frame = tstate->frame;
+
+    while (NULL != frame) {
+      int line = PyCode_Addr2Line(frame->f_code, frame->f_lasti);
+      std::string filename = THPUtils_unpackString(frame->f_code->co_filename);
+      std::string funcname = THPUtils_unpackString(frame->f_code->co_name);
+      stack_trace << filename << "(" << line << "): " << funcname << "\n";
+      frame = frame->f_back;
+    }
+  }
+  return stack_trace.str();
+}
+}  // namespace
+
 Node* recordTrace(std::string op, // TODO: make this a Symbol
                   at::ArrayRef<Variable> inputs,
                   at::ArrayRef<Variable> outputs) {
@@ -99,6 +126,9 @@ Node* recordTrace(std::string op, // TODO: make this a Symbol
   auto state_lock = state->lock();
 
   Node *n = graph->create(stringToSymbol(op));
+  auto sl = std::make_shared<SourceLocation>(getPythonInterpreterStackTrace());
+  n->setSourceLocation(sl);
+
   for (Variable input : inputs) {
     n->addInput(getValueTrace(state, input));
   }

torch/csrc/onnx/onnx.h

@@ -168,6 +168,21 @@ DEFINE_CONST(UINT64)
 DEFINE_CONST(COMPLEX64)
 DEFINE_CONST(COMPLEX128)
 #undef DEFINE_CONST
+
+#define DEFINE_CONST(C) \
+const auto a##C = onnx_AttributeProto_AttributeType_##C;
+DEFINE_CONST(FLOAT)
+DEFINE_CONST(INT)
+DEFINE_CONST(STRING)
+DEFINE_CONST(TENSOR)
+DEFINE_CONST(GRAPH)
+DEFINE_CONST(FLOATS)
+DEFINE_CONST(INTS)
+DEFINE_CONST(STRINGS)
+DEFINE_CONST(TENSORS)
+DEFINE_CONST(GRAPHS)
+#undef DEFINE_CONST
+
 // C++ wrappers which simulate the Google C++ Protobuf API
 //
 // These are NOT COMPLETE wrappers. If you find something is missing, add it!
@@ -270,6 +285,7 @@ public:
     proto.graphs  = list<GraphProto, onnx_GraphProto_fields>(&graphs);
   }
   void set_name(const std::string& s) { proto.name = string(&name, s); }
+  void set_type(onnx_AttributeProto_AttributeType t) { proto.has_type = true; proto.type = t; }
   void set_f(float f) { proto.has_f = true; proto.f = f; }
   void set_i(int64_t i) { proto.has_i = true; proto.i = i; }
   void set_s(std::string s_) { proto.s = string(&s, s_); }
@@ -290,6 +306,7 @@ public:
 class NodeProto : public MicroProto<onnx_NodeProto> {
 private:
   std::string op_type;
+  std::string doc_string;
   unique_vector<std::string> inputs;
   unique_vector<std::string> outputs;
   unique_vector<AttributeProto> attributes;
@@ -309,6 +326,7 @@ public:
     return ptr;
   }
   void set_op_type(const std::string& s) { proto.op_type= string(&op_type, s); }
+  void set_doc_string(const std::string& s) { proto.doc_string = string(&doc_string, s); }
 };
 
 class GraphProto : public MicroProto<onnx_GraphProto> {
@@ -349,6 +367,15 @@ public:
   }
 };
 
+class OperatorSetIdProto : public MicroProto<onnx_OperatorSetIdProto> {
+private:
+  std::string domain;
+public:
+  OperatorSetIdProto() : MicroProto(onnx_OperatorSetIdProto_init_default) {}
+  void set_domain(const std::string& s) { proto.domain = string(&domain, s); }
+  void set_version(int64_t v) { proto.has_version = true; proto.version = v; }
+};
+
 class ModelProto : public MicroProto<onnx_ModelProto> {
 private:
   std::string producer_name;
@@ -356,21 +383,26 @@ private:
   std::string domain;
   std::string doc_string;
   std::unique_ptr<GraphProto> graph;
+  unique_vector<OperatorSetIdProto> opset_import;
 public:
   ModelProto() : MicroProto(onnx_ModelProto_init_default) {
     proto.has_ir_version = true;
     proto.ir_version = onnx_Version_IR_VERSION;
-    proto.producer_name = string(&producer_name, "pytorch");
-    // TODO: stop hard-coding this
-    proto.producer_version = string(&producer_version, "0.2");
-    proto.domain = string(&domain, "com.facebook");
+    proto.opset_import = list<OperatorSetIdProto, onnx_OperatorSetIdProto_fields>(&opset_import);
   }
   void set_model_version(int64_t i) { proto.has_model_version = true; proto.model_version = i; }
   void set_doc_string(const std::string& s) { proto.doc_string = string(&doc_string, s); }
+  void set_producer_name(const std::string& s) { proto.producer_name = string(&producer_name, s); }
+  void set_producer_version(const std::string& s) { proto.producer_version = string(&producer_version, s); }
   GraphProto* mutable_graph() {
     proto.graph = msg<GraphProto, onnx_GraphProto_fields>(&graph);
     return graph.get();
   }
+  OperatorSetIdProto* add_opset_import() {
+    auto ptr = new OperatorSetIdProto();
+    opset_import.emplace_back(ptr);
+    return ptr;
+  }
 };
 
 }} // namespace torch::onnx

torch/csrc/onnx/onnx.pb.cpp

@@ -10,7 +10,7 @@
 
 
 
-const pb_field_t onnx_AttributeProto_fields[12] = {
+const pb_field_t onnx_AttributeProto_fields[13] = {
     PB_FIELD(  1, STRING  , OPTIONAL, CALLBACK, FIRST, onnx_AttributeProto, name, name, 0),
     PB_FIELD(  2, FLOAT   , OPTIONAL, STATIC  , OTHER, onnx_AttributeProto, f, name, 0),
     PB_FIELD(  3, INT64   , OPTIONAL, STATIC  , OTHER, onnx_AttributeProto, i, f, 0),
@@ -22,6 +22,7 @@ const pb_field_t onnx_AttributeProto_fields[12] = {
     PB_FIELD(  9, BYTES   , REPEATED, CALLBACK, OTHER, onnx_AttributeProto, strings, ints, 0),
     PB_FIELD( 10, MESSAGE , REPEATED, CALLBACK, OTHER, onnx_AttributeProto, tensors, strings, &onnx_TensorProto_fields),
     PB_FIELD( 11, MESSAGE , REPEATED, CALLBACK, OTHER, onnx_AttributeProto, graphs, tensors, &onnx_GraphProto_fields),
+    PB_FIELD( 20, UENUM   , OPTIONAL, STATIC  , OTHER, onnx_AttributeProto, type, graphs, 0),
     PB_LAST_FIELD
 };
 
@@ -31,17 +32,18 @@ const pb_field_t onnx_ValueInfoProto_fields[3] = {
     PB_LAST_FIELD
 };
 
-const pb_field_t onnx_NodeProto_fields[7] = {
+const pb_field_t onnx_NodeProto_fields[8] = {
     PB_FIELD(  1, STRING  , REPEATED, CALLBACK, FIRST, onnx_NodeProto, input, input, 0),
     PB_FIELD(  2, STRING  , REPEATED, CALLBACK, OTHER, onnx_NodeProto, output, input, 0),
     PB_FIELD(  3, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_NodeProto, name, output, 0),
     PB_FIELD(  4, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_NodeProto, op_type, name, 0),
     PB_FIELD(  5, MESSAGE , REPEATED, CALLBACK, OTHER, onnx_NodeProto, attribute, op_type, &onnx_AttributeProto_fields),
     PB_FIELD(  6, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_NodeProto, doc_string, attribute, 0),
+    PB_FIELD(  7, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_NodeProto, domain, doc_string, 0),
     PB_LAST_FIELD
 };
 
-const pb_field_t onnx_ModelProto_fields[8] = {
+const pb_field_t onnx_ModelProto_fields[9] = {
     PB_FIELD(  1, INT64   , OPTIONAL, STATIC  , FIRST, onnx_ModelProto, ir_version, ir_version, 0),
     PB_FIELD(  2, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_ModelProto, producer_name, ir_version, 0),
     PB_FIELD(  3, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_ModelProto, producer_version, producer_name, 0),
@@ -49,6 +51,7 @@ const pb_field_t onnx_ModelProto_fields[8] = {
     PB_FIELD(  5, INT64   , OPTIONAL, STATIC  , OTHER, onnx_ModelProto, model_version, domain, 0),
     PB_FIELD(  6, STRING  , OPTIONAL, CALLBACK, OTHER, onnx_ModelProto, doc_string, model_version, 0),
     PB_FIELD(  7, MESSAGE , OPTIONAL, CALLBACK, OTHER, onnx_ModelProto, graph, doc_string, &onnx_GraphProto_fields),
+    PB_FIELD(  8, MESSAGE , REPEATED, CALLBACK, OTHER, onnx_ModelProto, opset_import, graph, &onnx_OperatorSetIdProto_fields),
     PB_LAST_FIELD
 };
 
@@ -120,6 +123,13 @@ const pb_field_t onnx_TypeProto_SparseTensorTypeProto_fields[3] = {
     PB_LAST_FIELD
 };
 
+const pb_field_t onnx_OperatorSetIdProto_fields[3] = {
+    PB_FIELD(  1, STRING  , OPTIONAL, CALLBACK, FIRST, onnx_OperatorSetIdProto, domain, domain, 0),
+    PB_FIELD(  2, INT64   , OPTIONAL, STATIC  , OTHER, onnx_OperatorSetIdProto, version, domain, 0),
+    PB_LAST_FIELD
+};
+
+
 
 
 
@@ -132,7 +142,7 @@ const pb_field_t onnx_TypeProto_SparseTensorTypeProto_fields[3] = {
  * numbers or field sizes that are larger than what can fit in 8 or 16 bit
  * field descriptors.
  */
-PB_STATIC_ASSERT((pb_membersize(onnx_TensorProto, segment) < 65536 && pb_membersize(onnx_SparseTensorProto, indices) < 65536 && pb_membersize(onnx_SparseTensorProto, values) < 65536 && pb_membersize(onnx_TypeProto, sparse_tensor_type) < 65536 && pb_membersize(onnx_TypeProto_SparseTensorTypeProto, shape) < 65536), YOU_MUST_DEFINE_PB_FIELD_32BIT_FOR_MESSAGES_onnx_AttributeProto_onnx_ValueInfoProto_onnx_NodeProto_onnx_ModelProto_onnx_GraphProto_onnx_TensorProto_onnx_TensorProto_Segment_onnx_SparseTensorProto_onnx_TypeProto_onnx_TypeProto_TensorShapeProto_onnx_TypeProto_TensorShapeProto_Dimension_onnx_TypeProto_TensorTypeProto_onnx_TypeProto_SparseTensorTypeProto)
+PB_STATIC_ASSERT((pb_membersize(onnx_TensorProto, segment) < 65536 && pb_membersize(onnx_SparseTensorProto, indices) < 65536 && pb_membersize(onnx_SparseTensorProto, values) < 65536 && pb_membersize(onnx_TypeProto, sparse_tensor_type) < 65536 && pb_membersize(onnx_TypeProto_SparseTensorTypeProto, shape) < 65536), YOU_MUST_DEFINE_PB_FIELD_32BIT_FOR_MESSAGES_onnx_AttributeProto_onnx_ValueInfoProto_onnx_NodeProto_onnx_ModelProto_onnx_GraphProto_onnx_TensorProto_onnx_TensorProto_Segment_onnx_SparseTensorProto_onnx_TypeProto_onnx_TypeProto_TensorShapeProto_onnx_TypeProto_TensorShapeProto_Dimension_onnx_TypeProto_TensorTypeProto_onnx_TypeProto_SparseTensorTypeProto_onnx_OperatorSetIdProto)
 #endif
 
 #if !defined(PB_FIELD_16BIT) && !defined(PB_FIELD_32BIT)
@@ -143,7 +153,7 @@ PB_STATIC_ASSERT((pb_membersize(onnx_TensorProto, segment) < 65536 && pb_members
  * numbers or field sizes that are larger than what can fit in the default
  * 8 bit descriptors.
  */
-PB_STATIC_ASSERT((pb_membersize(onnx_TensorProto, segment) < 256 && pb_membersize(onnx_SparseTensorProto, indices) < 256 && pb_membersize(onnx_SparseTensorProto, values) < 256 && pb_membersize(onnx_TypeProto, sparse_tensor_type) < 256 && pb_membersize(onnx_TypeProto_SparseTensorTypeProto, shape) < 256), YOU_MUST_DEFINE_PB_FIELD_16BIT_FOR_MESSAGES_onnx_AttributeProto_onnx_ValueInfoProto_onnx_NodeProto_onnx_ModelProto_onnx_GraphProto_onnx_TensorProto_onnx_TensorProto_Segment_onnx_SparseTensorProto_onnx_TypeProto_onnx_TypeProto_TensorShapeProto_onnx_TypeProto_TensorShapeProto_Dimension_onnx_TypeProto_TensorTypeProto_onnx_TypeProto_SparseTensorTypeProto)
+PB_STATIC_ASSERT((pb_membersize(onnx_TensorProto, segment) < 256 && pb_membersize(onnx_SparseTensorProto, indices) < 256 && pb_membersize(onnx_SparseTensorProto, values) < 256 && pb_membersize(onnx_TypeProto, sparse_tensor_type) < 256 && pb_membersize(onnx_TypeProto_SparseTensorTypeProto, shape) < 256), YOU_MUST_DEFINE_PB_FIELD_16BIT_FOR_MESSAGES_onnx_AttributeProto_onnx_ValueInfoProto_onnx_NodeProto_onnx_ModelProto_onnx_GraphProto_onnx_TensorProto_onnx_TensorProto_Segment_onnx_SparseTensorProto_onnx_TypeProto_onnx_TypeProto_TensorShapeProto_onnx_TypeProto_TensorShapeProto_Dimension_onnx_TypeProto_TensorTypeProto_onnx_TypeProto_SparseTensorTypeProto_onnx_OperatorSetIdProto)
 #endif
 
 

torch/csrc/onnx/onnx.pb.h

@@ -16,12 +16,31 @@ extern "C" {
 
 /* Enum definitions */
 typedef enum _onnx_Version {
-    onnx_Version_IR_VERSION = 1
+    onnx_Version__START_VERSION = 0,
+    onnx_Version_IR_VERSION_2017_10_10 = 1,
+    onnx_Version_IR_VERSION = 2
 } onnx_Version;
-#define _onnx_Version_MIN onnx_Version_IR_VERSION
+#define _onnx_Version_MIN onnx_Version__START_VERSION
 #define _onnx_Version_MAX onnx_Version_IR_VERSION
 #define _onnx_Version_ARRAYSIZE ((onnx_Version)(onnx_Version_IR_VERSION+1))
 
+typedef enum _onnx_AttributeProto_AttributeType {
+    onnx_AttributeProto_AttributeType_UNDEFINED = 0,
+    onnx_AttributeProto_AttributeType_FLOAT = 1,
+    onnx_AttributeProto_AttributeType_INT = 2,
+    onnx_AttributeProto_AttributeType_STRING = 3,
+    onnx_AttributeProto_AttributeType_TENSOR = 4,
+    onnx_AttributeProto_AttributeType_GRAPH = 5,
+    onnx_AttributeProto_AttributeType_FLOATS = 6,
+    onnx_AttributeProto_AttributeType_INTS = 7,
+    onnx_AttributeProto_AttributeType_STRINGS = 8,
+    onnx_AttributeProto_AttributeType_TENSORS = 9,
+    onnx_AttributeProto_AttributeType_GRAPHS = 10
+} onnx_AttributeProto_AttributeType;
+#define _onnx_AttributeProto_AttributeType_MIN onnx_AttributeProto_AttributeType_UNDEFINED
+#define _onnx_AttributeProto_AttributeType_MAX onnx_AttributeProto_AttributeType_GRAPHS
+#define _onnx_AttributeProto_AttributeType_ARRAYSIZE ((onnx_AttributeProto_AttributeType)(onnx_AttributeProto_AttributeType_GRAPHS+1))
+
 typedef enum _onnx_TensorProto_DataType {
     onnx_TensorProto_DataType_UNDEFINED = 0,
     onnx_TensorProto_DataType_FLOAT = 1,
@@ -63,6 +82,7 @@ typedef struct _onnx_NodeProto {
     pb_callback_t op_type;
     pb_callback_t attribute;
     pb_callback_t doc_string;
+    pb_callback_t domain;
 /* @@protoc_insertion_point(struct:onnx_NodeProto) */
 } onnx_NodeProto;
 
@@ -91,6 +111,8 @@ typedef struct _onnx_AttributeProto {
     pb_callback_t strings;
     pb_callback_t tensors;
     pb_callback_t graphs;
+    bool has_type;
+    onnx_AttributeProto_AttributeType type;
 /* @@protoc_insertion_point(struct:onnx_AttributeProto) */
 } onnx_AttributeProto;
 
@@ -104,9 +126,17 @@ typedef struct _onnx_ModelProto {
     int64_t model_version;
     pb_callback_t doc_string;
     pb_callback_t graph;
+    pb_callback_t opset_import;
 /* @@protoc_insertion_point(struct:onnx_ModelProto) */
 } onnx_ModelProto;
 
+typedef struct _onnx_OperatorSetIdProto {
+    pb_callback_t domain;
+    bool has_version;
+    int64_t version;
+/* @@protoc_insertion_point(struct:onnx_OperatorSetIdProto) */
+} onnx_OperatorSetIdProto;
+
 typedef struct _onnx_TensorProto_Segment {
     bool has_begin;
     int64_t begin;
@@ -173,10 +203,10 @@ typedef struct _onnx_SparseTensorProto {
 /* Default values for struct fields */
 
 /* Initializer values for message structs */
-#define onnx_AttributeProto_init_default         {{{NULL}, NULL}, false, 0, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_AttributeProto_init_default         {{{NULL}, NULL}, false, 0, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, (onnx_AttributeProto_AttributeType)0}
 #define onnx_ValueInfoProto_init_default         {{{NULL}, NULL}, {{NULL}, NULL}}
-#define onnx_NodeProto_init_default              {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
-#define onnx_ModelProto_init_default             {false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, 0, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_NodeProto_init_default              {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_ModelProto_init_default             {false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_GraphProto_init_default             {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_TensorProto_init_default            {{{NULL}, NULL}, false, (onnx_TensorProto_DataType)0, false, onnx_TensorProto_Segment_init_default, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_TensorProto_Segment_init_default    {false, 0, false, 0}
@@ -186,10 +216,11 @@ typedef struct _onnx_SparseTensorProto {
 #define onnx_TypeProto_TensorShapeProto_Dimension_init_default {false, 0, {{NULL}, NULL}}
 #define onnx_TypeProto_TensorTypeProto_init_default {false, (onnx_TensorProto_DataType)0, {{NULL}, NULL}}
 #define onnx_TypeProto_SparseTensorTypeProto_init_default {false, (onnx_TensorProto_DataType)0, false, onnx_TypeProto_TensorShapeProto_init_default}
-#define onnx_AttributeProto_init_zero            {{{NULL}, NULL}, false, 0, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_OperatorSetIdProto_init_default     {{{NULL}, NULL}, false, 0}
+#define onnx_AttributeProto_init_zero            {{{NULL}, NULL}, false, 0, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, (onnx_AttributeProto_AttributeType)0}
 #define onnx_ValueInfoProto_init_zero            {{{NULL}, NULL}, {{NULL}, NULL}}
-#define onnx_NodeProto_init_zero                 {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
-#define onnx_ModelProto_init_zero                {false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, 0, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_NodeProto_init_zero                 {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
+#define onnx_ModelProto_init_zero                {false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, false, 0, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_GraphProto_init_zero                {{{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_TensorProto_init_zero               {{{NULL}, NULL}, false, (onnx_TensorProto_DataType)0, false, onnx_TensorProto_Segment_init_zero, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}, {{NULL}, NULL}}
 #define onnx_TensorProto_Segment_init_zero       {false, 0, false, 0}
@@ -199,6 +230,7 @@ typedef struct _onnx_SparseTensorProto {
 #define onnx_TypeProto_TensorShapeProto_Dimension_init_zero {false, 0, {{NULL}, NULL}}
 #define onnx_TypeProto_TensorTypeProto_init_zero {false, (onnx_TensorProto_DataType)0, {{NULL}, NULL}}
 #define onnx_TypeProto_SparseTensorTypeProto_init_zero {false, (onnx_TensorProto_DataType)0, false, onnx_TypeProto_TensorShapeProto_init_zero}
+#define onnx_OperatorSetIdProto_init_zero        {{{NULL}, NULL}, false, 0}
 
 /* Field tags (for use in manual encoding/decoding) */
 #define onnx_GraphProto_node_tag                 1
@@ -212,12 +244,14 @@ typedef struct _onnx_SparseTensorProto {
 #define onnx_NodeProto_output_tag                2
 #define onnx_NodeProto_name_tag                  3
 #define onnx_NodeProto_op_type_tag               4
+#define onnx_NodeProto_domain_tag                7
 #define onnx_NodeProto_attribute_tag             5
 #define onnx_NodeProto_doc_string_tag            6
 #define onnx_TypeProto_TensorShapeProto_dim_tag  1
 #define onnx_ValueInfoProto_name_tag             1
 #define onnx_ValueInfoProto_type_tag             2
 #define onnx_AttributeProto_name_tag             1
+#define onnx_AttributeProto_type_tag             20
 #define onnx_AttributeProto_f_tag                2
 #define onnx_AttributeProto_i_tag                3
 #define onnx_AttributeProto_s_tag                4
@@ -229,12 +263,15 @@ typedef struct _onnx_SparseTensorProto {
 #define onnx_AttributeProto_tensors_tag          10
 #define onnx_AttributeProto_graphs_tag           11
 #define onnx_ModelProto_ir_version_tag           1
+#define onnx_ModelProto_opset_import_tag         8
 #define onnx_ModelProto_producer_name_tag        2
 #define onnx_ModelProto_producer_version_tag     3
 #define onnx_ModelProto_domain_tag               4
 #define onnx_ModelProto_model_version_tag        5
 #define onnx_ModelProto_doc_string_tag           6
 #define onnx_ModelProto_graph_tag                7
+#define onnx_OperatorSetIdProto_domain_tag       1
+#define onnx_OperatorSetIdProto_version_tag      2
 #define onnx_TensorProto_Segment_begin_tag       1
 #define onnx_TensorProto_Segment_end_tag         2
 #define onnx_TypeProto_SparseTensorTypeProto_elem_type_tag 1
@@ -261,10 +298,10 @@ typedef struct _onnx_SparseTensorProto {
 #define onnx_SparseTensorProto_values_tag        3
 
 /* Struct field encoding specification for nanopb */
-extern const pb_field_t onnx_AttributeProto_fields[12];
+extern const pb_field_t onnx_AttributeProto_fields[13];
 extern const pb_field_t onnx_ValueInfoProto_fields[3];
-extern const pb_field_t onnx_NodeProto_fields[7];
-extern const pb_field_t onnx_ModelProto_fields[8];
+extern const pb_field_t onnx_NodeProto_fields[8];
+extern const pb_field_t onnx_ModelProto_fields[9];
 extern const pb_field_t onnx_GraphProto_fields[8];
 extern const pb_field_t onnx_TensorProto_fields[12];
 extern const pb_field_t onnx_TensorProto_Segment_fields[3];
@@ -274,6 +311,7 @@ extern const pb_field_t onnx_TypeProto_TensorShapeProto_fields[2];
 extern const pb_field_t onnx_TypeProto_TensorShapeProto_Dimension_fields[3];
 extern const pb_field_t onnx_TypeProto_TensorTypeProto_fields[3];
 extern const pb_field_t onnx_TypeProto_SparseTensorTypeProto_fields[3];
+extern const pb_field_t onnx_OperatorSetIdProto_fields[3];
 
 /* Maximum encoded size of messages (where known) */
 /* onnx_AttributeProto_size depends on runtime parameters */
@@ -289,6 +327,7 @@ extern const pb_field_t onnx_TypeProto_SparseTensorTypeProto_fields[3];
 /* onnx_TypeProto_TensorShapeProto_Dimension_size depends on runtime parameters */
 /* onnx_TypeProto_TensorTypeProto_size depends on runtime parameters */
 #define onnx_TypeProto_SparseTensorTypeProto_size (8 + onnx_TypeProto_TensorShapeProto_size)
+/* onnx_OperatorSetIdProto_size depends on runtime parameters */
 
 /* Message IDs (where set with "msgid" option) */
 #ifdef PB_MSGID

torch/cuda/__init__.py

@@ -14,6 +14,7 @@ import ctypes
 import os
 import torch
 import traceback
+import warnings
 from torch._six import raise_from
 from multiprocessing.util import register_after_fork as _register_after_fork
 
@@ -65,11 +66,29 @@ http://www.nvidia.com/Download/index.aspx""")
 The NVIDIA driver on your system is too old (found version {}).
 Please update your GPU driver by downloading and installing a new
 version from the URL: http://www.nvidia.com/Download/index.aspx
-Alternatively, go to: https://pytorch.org/binaries to install
+Alternatively, go to: http://pytorch.org to install
 a PyTorch version that has been compiled with your version
 of the CUDA driver.""".format(str(torch._C._cuda_getDriverVersion())))
 
 
+def _check_capability():
+    error_str = """
+    Found GPU%d %s which requires CUDA_VERSION >= %d for
+     optimal performance and fast startup time, but your PyTorch was compiled
+     with CUDA_VERSION %d. Please install the correct PyTorch binary
+     using instructions from http://pytorch.org
+    """
+
+    CUDA_VERSION = torch._C._cuda_getCompiledVersion()
+    for d in range(device_count()):
+        major = get_device_capability(d)[0]
+        name = get_device_name(d)
+        if CUDA_VERSION < 8000 and major >= 6:
+            warnings.warn(error_str % (d, name, 8000, CUDA_VERSION))
+        elif CUDA_VERSION < 9000 and major >= 7:
+            warnings.warn(error_str % (d, name, 8000, CUDA_VERSION))
+
+
 def _lazy_call(callable):
     if _initialized:
         callable()
@@ -77,6 +96,8 @@ def _lazy_call(callable):
         # Don't store the actual traceback to avoid memory cycle
         _queued_calls.append((callable, traceback.format_stack()))
 
+_lazy_call(_check_capability)
+
 
 class DeferredCudaCallError(Exception):
     pass
@@ -213,6 +234,19 @@ def get_device_name(device):
         return torch._C._cuda_getDeviceName(device)
 
 
+def get_device_capability(device):
+    """Gets the cuda capability of a device.
+
+    Arguments:
+        device (int): device for which to return the name. This function is a
+            no-op if this argument is negative.
+    Returns:
+        tuple(int, int): the major and minor cuda capability of the device
+    """
+    if device >= 0:
+        return torch._C._cuda_getDeviceCapability(device)
+
+
 @contextlib.contextmanager
 def stream(stream):
     """Context-manager that selects a given stream.
@@ -267,6 +301,13 @@ def current_blas_handle():
     return torch._C._cuda_getCurrentBlasHandle()
 
 
+def empty_cache():
+    """Releases all unoccupied cached memory currently held by the caching
+    allocator so that those can be used in other GPU application and visible in
+    `nvidia-smi`."""
+    return torch._C._cuda_emptyCache()
+
+
 def _host_allocator():
     _lazy_init()
     return torch._C._cuda_cudaHostAllocator()

torch/cuda/streams.py

@@ -107,10 +107,10 @@ class Event(object):
 
     Arguments:
         enable_timing (bool): indicates if the event should measure time
-            (default: False)
-        blocking (bool): if true, :meth:`wait` will be blocking (default: False)
-        interprocess (bool): if true, the event can be shared between processes
-            (default: False)
+            (default: ``False``)
+        blocking (bool): if ``True``, :meth:`wait` will be blocking (default: ``False``)
+        interprocess (bool): if ``True``, the event can be shared between processes
+            (default: ``False``)
     """
 
     DEFAULT = 0x0

torch/distributions.py

@@ -1,17 +1,32 @@
-"""
+r"""
 The ``distributions`` package contains parameterizable probability distributions
 and sampling functions.
 
-The :meth:`log_prob` method is useful for policy gradient based methods. If the
-parameters of the distribution are differentiable, then the result of ``log_prob``
-is also differentiable.
+Policy gradient methods can be implemented using the
+:meth:`~torch.distributions.Distribution.log_prob` method, when the probability
+density function is differentiable with respect to its parameters. A basic
+method is the REINFORCE rule:
 
-Example::
+.. math::
 
-    probs = network(input)
-    m = Multinomial(probs)
+    \Delta\theta  = \alpha r \frac{\partial\log p(a|\pi^\theta(s))}{\partial\theta}
+
+where :math:`\theta` are the parameters, :math:`\alpha` is the learning rate,
+:math:`r` is the reward and :math:`p(a|\pi^\theta(s))` is the probability of
+taking action :math:`a` in state :math:`s` given policy :math:`\pi^\theta`.
+
+In practice we would sample an action from the output of a network, apply this
+action in an environment, and then use ``log_prob`` to construct an equivalent
+loss function. Note that we use a negative because optimisers use gradient
+descent, whilst the rule above assumes gradient ascent. With a categorical
+policy, the code for implementing REINFORCE would be as follows::
+
+    probs = policy_network(state)
+    # NOTE: this is equivalent to what used to be called multinomial
+    m = Categorical(probs)
     action = m.sample()
-    loss = -m.log_prob(action) * get_reward(env, action)
+    next_state, reward = env.step(action)
+    loss = -m.log_prob(action) * reward
     loss.backward()
 """
 import math
@@ -19,7 +34,7 @@ from numbers import Number
 import torch
 
 
-__all__ = ['Distribution', 'Bernoulli', 'Multinomial', 'Normal']
+__all__ = ['Distribution', 'Bernoulli', 'Categorical', 'Normal']
 
 
 class Distribution(object):
@@ -87,9 +102,12 @@ class Bernoulli(Distribution):
         return log_pmf.gather(0, value.unsqueeze(0).long()).squeeze(0)
 
 
-class Multinomial(Distribution):
+class Categorical(Distribution):
     r"""
-    Creates a multinomial distribution parameterized by `probs`.
+    Creates a categorical distribution parameterized by `probs`.
+
+    .. note::
+        It is equivalent to the distribution that ``multinomial()`` samples from.
 
     Samples are integers from `0 ... K-1` where `K` is probs.size(-1).
 
@@ -102,7 +120,7 @@ class Multinomial(Distribution):
 
     Example::
 
-        >>> m = Multinomial(torch.Tensor([ 0.25, 0.25, 0.25, 0.25 ]))
+        >>> m = Categorical(torch.Tensor([ 0.25, 0.25, 0.25, 0.25 ]))
         >>> m.sample()  # equal probability of 0, 1, 2, 3
          3
         [torch.LongTensor of size 1]

torch/jit/__init__.py

@@ -69,10 +69,10 @@ def compile(arg=None, **kwargs):
             (as we always wait to see all derivatives before compiling.)
             Default: 1 (i.e., we will compile forwards and backwards, but not
             double-backwards).
-        optimize (bool, optional): whether or not to apply optimizations.  Default: True.
+        optimize (bool, optional): whether or not to apply optimizations.  Default: ``True``.
 
     Debug arguments:
-        time (bool, optional): if True, whenever we execute the model in question, we
+        time (bool, optional): if ``True``, whenever we execute the model in question, we
             will also print out some timing information for how long the model
             took to execute.  At the moment, there are three types of timings we
             emit:
@@ -87,10 +87,10 @@ def compile(arg=None, **kwargs):
                 - optimized: the time it took to execute the optimized model.
 
             At the moment, all of these timings are for the forward pass only.
-            Default: False.
-        enabled (bool, optional): if False, compilation is disabled and you
+            Default: ``False``.
+        enabled (bool, optional): if ``False``, compilation is disabled and you
             will get back your original model.  This is a convenient way to
-            disable tracing without having to delete the annotation. Default: True.
+            disable tracing without having to delete the annotation. Default: ``True``.
 
     Example: Compile as class decorator.
 
@@ -396,6 +396,10 @@ class _CompiledMixin(object):
         # TODO: Figure out how to call parent destructor, if there is one.
         # Apparently, this is buggy:
         #     https://stackoverflow.com/questions/22972720/python-cant-invoke-parent-class-destructor-with-super
+        # NB: Have to mangle this by hand!
+        if not (hasattr(self, '_CompiledMixin__misses') and hasattr(self, '_CompiledMixin___hits')):
+            # Probably died during construction
+            return
         if self.__misses != 0 and self.__hits == 0:
             warnings.warn("{} was marked with JIT and invoked {} times, "
                           "but we never successfully used compiled code."

torch/legacy/nn/DistKLDivCriterion.py

@@ -18,18 +18,22 @@ class DistKLDivCriterion(Criterion):
             input,
             target,
             self.output_tensor,
-            self.sizeAverage
+            self.sizeAverage,
+            True,  # reduce
         )
         self.output = self.output_tensor[0]
         return self.output
 
     def updateGradInput(self, input, target):
         assert input.is_same_size(target)
+        implicit_gradOutput = torch.ones(1).type_as(input)
         self._backend.DistKLDivCriterion_updateGradInput(
             self._backend.library_state,
             input,
             target,
+            implicit_gradOutput,
             self.gradInput,
-            self.sizeAverage
+            self.sizeAverage,
+            True,  # reduce
         )
         return self.gradInput

torch/legacy/nn/ELU.py

@@ -29,7 +29,6 @@ class ELU(Module):
     def updateGradInput(self, input, gradOutput):
         self._backend.ELU_updateGradInput(
             self._backend.library_state,
-            input,
             gradOutput,
             self.gradInput,
             self.output,

torch/lib/ATen/CMakeLists.txt

@@ -66,6 +66,7 @@ IF ($ENV{TH_BINARY_BUILD})
   IF (UNIX AND NOT APPLE)
     # hiding statically linked library symbols, this flag is not available for the linker under MACOSX
     SET(CMAKE_CXX_FLAGS "-Wl,--exclude-libs,libstdc++.a ${CMAKE_CXX_FLAGS}")
+    set (CMAKE_SHARED_LINKER_FLAGS "-Wl,--version-script=${CMAKE_CURRENT_SOURCE_DIR}/../../../tools/pytorch.version")
   ENDIF(UNIX AND NOT APPLE)
 ENDIF()
 

torch/lib/ATen/Context.h

@@ -17,8 +17,15 @@ public:
   Type & getType(Backend p, ScalarType s) {
     initCUDAIfNeeded(p);
     auto & type = type_registry[static_cast<int>(p)][static_cast<int>(s)];
-    if(!type)
+
+    if(!type) {
+      // there is only a single Undefined Type.
+      if (p == Backend::Undefined || s == ScalarType::Undefined) {
+        auto & undef = type_registry[static_cast<int>(Backend::Undefined)][static_cast<int>(ScalarType::Undefined)];
+        if (undef) return *undef;
+      }
       runtime_error("%s%sType is not enabled.",toString(p),toString(s));
+    }
     return *type;
   }
   Generator & defaultGenerator(Backend p) {

torch/lib/ATen/DLConvertor.cpp

@@ -36,6 +36,8 @@ static DLDataType getDLDataType(const Type& type) {
     case ScalarType::Half:
       dtype.code = DLDataTypeCode::kFloat;
       break;
+    case ScalarType::Undefined:
+      throw std::logic_error("Undefined is not a valid ScalarType");
     case ScalarType::NumOptions:
       throw std::logic_error("NumOptions is not a valid ScalarType");
   }

torch/lib/ATen/Declarations.cwrap

@@ -579,13 +579,22 @@
     - CPU
     - CUDA
   return: argument 0
-  arguments:
-    - arg: THTensor* result
-      output: True
-    - accreal start
-    - accreal end
-    - arg: accreal step
-      default: 1
+  options:
+    - cname: arange
+      arguments:
+        - arg: THTensor* result
+          output: True
+        - accreal start
+        - accreal end
+        - arg: accreal step
+          default: 1
+    - cname: arange
+      arguments:
+        - arg: THTensor* result
+          output: True
+        - CONSTANT 0
+        - accreal end
+        - CONSTANT 1
 ]]
 [[
   name: scatter_

torch/lib/ATen/ExpandUtils.h

@@ -1,10 +1,20 @@
 #pragma once
 
 #include "ATen/Tensor.h"
+#include <functional>
 #include <sstream>
 
 namespace at {
 
+// avoid copy-construction of Tensor by using a reference_wrapper.
+inline void check_defined(std::initializer_list<std::reference_wrapper<const Tensor>> tensors, const char *api_name) {
+  for (auto& t : tensors) {
+    if (!t.get().defined()) {
+      runtime_error("%s(...) called with an undefined Tensor", api_name);
+    }
+  }
+}
+
 inline std::tuple<Tensor> expand_inplace(const Tensor &tensor, const Tensor &to_expand) {
   if (tensor.sizes().equals(to_expand.sizes())) {
     return std::make_tuple(to_expand);
@@ -13,6 +23,11 @@ inline std::tuple<Tensor> expand_inplace(const Tensor &tensor, const Tensor &to_
   return std::make_tuple(to_expand.expand(tensor.sizes()));
 }
 
+inline std::tuple<Tensor> expand_inplace(const Tensor &tensor, const Tensor &to_expand, const char *api_name) {
+  check_defined({tensor, to_expand}, api_name);
+  return expand_inplace(tensor, to_expand);
+}
+
 inline std::tuple<Tensor, Tensor> expand_inplace(const Tensor &tensor, const Tensor &to_expand1, const Tensor &to_expand2) {
   if (tensor.sizes().equals(to_expand1.sizes()) && tensor.sizes().equals((to_expand2.sizes()))) {
     return std::make_tuple(to_expand1, to_expand2);
@@ -21,6 +36,12 @@ inline std::tuple<Tensor, Tensor> expand_inplace(const Tensor &tensor, const Ten
   return std::make_tuple(to_expand1.expand(tensor.sizes()), to_expand2.expand(tensor.sizes()));
 }
 
+inline std::tuple<Tensor, Tensor> expand_inplace(const Tensor &tensor, const Tensor &to_expand1, const Tensor &to_expand2,
+                                                 const char *api_name) {
+  check_defined({tensor, to_expand1, to_expand2}, api_name);
+  return expand_inplace(tensor, to_expand1, to_expand2);
+}
+
 inline std::vector<int64_t> infer_size2(IntList a, IntList b) {
   auto dimsA = a.size();
   auto dimsB = b.size();
@@ -55,9 +76,14 @@ inline std::tuple<Tensor, Tensor> expand_outplace(const Tensor &to_expand1, cons
   return std::make_tuple(to_expand1.expand(expanded_size), to_expand2.expand(expanded_size));
 }
 
-std::tuple<Tensor, Tensor, Tensor> expand_outplace(const Tensor &to_expand1,
-                                                   const Tensor &to_expand2,
-                                                   const Tensor &to_expand3) {
+inline std::tuple<Tensor, Tensor> expand_outplace(const Tensor &to_expand1, const Tensor &to_expand2, const char *api_name) {
+  check_defined({to_expand1, to_expand2}, api_name);
+  return expand_outplace(to_expand1, to_expand2);
+}
+
+inline std::tuple<Tensor, Tensor, Tensor> expand_outplace(const Tensor &to_expand1,
+                                                          const Tensor &to_expand2,
+                                                          const Tensor &to_expand3) {
   if (to_expand1.sizes().equals(to_expand2.sizes()) && to_expand1.sizes().equals(to_expand3.sizes())) {
     return std::make_tuple(to_expand1, to_expand2, to_expand3);
   }
@@ -67,6 +93,14 @@ std::tuple<Tensor, Tensor, Tensor> expand_outplace(const Tensor &to_expand1,
   return std::make_tuple(to_expand1.expand(expanded_size), to_expand2.expand(expanded_size), to_expand3.expand(expanded_size));
 }
 
+inline std::tuple<Tensor, Tensor, Tensor> expand_outplace(const Tensor &to_expand1,
+                                                          const Tensor &to_expand2,
+                                                          const Tensor &to_expand3,
+                                                          const char *api_name) {
+  check_defined({to_expand1, to_expand2, to_expand3}, api_name);
+  return expand_outplace(to_expand1, to_expand2, to_expand3);
+}
+
 inline std::tuple<Tensor> expand_size(const Tensor &to_expand, IntList sizes) {
   if(to_expand.sizes().equals(sizes)) {
     return std::make_tuple(to_expand);
@@ -75,4 +109,9 @@ inline std::tuple<Tensor> expand_size(const Tensor &to_expand, IntList sizes) {
   return std::make_tuple(to_expand.expand(sizes));
 }
 
+inline std::tuple<Tensor> expand_size(const Tensor &to_expand, IntList sizes, const char *api_name) {
+  check_defined({to_expand}, api_name);
+  return expand_size(to_expand, sizes);
+}
+
 }

torch/lib/ATen/Local.cwrap

@@ -128,6 +128,24 @@
     ${THTensor}_setStorage(${state,}result_->tensor, self_->tensor->storage, self_->tensor->storageOffset, size_, stride_);
 ]]
 
+[[
+  name: as_strided_
+  variants: [method,function]
+  return: argument 0
+  arguments:
+    - THTensor* self
+    - THSize* size
+    - THStride* stride
+    - arg: int64_t storage_offset
+      default: -1
+  aten_custom_call: |
+    if (storage_offset == -1) {
+      storage_offset = self_->tensor->storageOffset;
+    }
+    ${THTensor}_setStorage(${state,}self_->tensor, self_->tensor->storage, storage_offset, size_, stride_);
+    self_->maybeScalar(size.size() == 0);
+]]
+
 [[
   name: cat
   cname: catArray

torch/lib/ATen/Scalar.h

@@ -23,7 +23,7 @@ public:
 
   explicit Scalar(const detail::TensorBase & t)
   : tag(Tag::HAS_t), t(t) {
-    AT_ASSERT(t.pImpl, "Attempting to create a Scalar from an undefined tensor");
+    AT_ASSERT(t.defined(), "Attempting to create a Scalar from an undefined tensor");
     AT_ASSERT(t.dim() == 0, "Attempting to create a Scalar from a %d dim tensor", t.dim());
   }
 

torch/lib/ATen/ScalarType.h

@@ -23,6 +23,7 @@ enum class ScalarType {
   n,
   AT_FORALL_SCALAR_TYPES(DEFINE_ENUM)
 #undef DEFINE_ENUM
+  Undefined,
   NumOptions
 };
 
@@ -31,6 +32,7 @@ enum class Backend {
   CUDA,
   SparseCPU,
   SparseCUDA,
+  Undefined,
   NumOptions
 };
 
@@ -62,7 +64,7 @@ static inline const char * toString(ScalarType t) {
   switch(t) {
     AT_FORALL_SCALAR_TYPES(DEFINE_CASE)
     default:
-      return "UNKNOWN_SCALAR_TYPE";
+      return "UNKNOWN_SCALAR";
   }
 #undef DEFINE_CASE
 }

torch/lib/ATen/TensorBase.h

@@ -1,29 +1,32 @@
 #pragma once
 
 #include "ATen/TensorImpl.h"
+#include "ATen/UndefinedTensor.h"
 
 namespace at { namespace detail {
 
 // TensorBase is the base class for Tensor which handles the reference counting
 struct TensorBase {
-  TensorBase()
-  : pImpl(nullptr) {}
+  TensorBase(): TensorBase(UndefinedTensor::singleton(), false) {}
   TensorBase(TensorImpl * self, bool retain)
   : pImpl(self) {
-    if(pImpl != nullptr && retain)
+    if (pImpl == nullptr) {
+      throw std::runtime_error("TensorBase with nullptr not supported");
+    }
+    if(retain && pImpl != UndefinedTensor::singleton())
       pImpl->retain();
   }
   TensorBase(const TensorBase & rhs)
   : pImpl(rhs.pImpl) {
-    if(pImpl != nullptr)
+    if (pImpl != UndefinedTensor::singleton())
       pImpl->retain();
   }
   TensorBase(TensorBase && rhs) noexcept
   : pImpl(rhs.pImpl) {
-    rhs.pImpl = nullptr;
+    rhs.pImpl = UndefinedTensor::singleton();
   }
   ~TensorBase() {
-    if(pImpl != nullptr)
+    if (pImpl != UndefinedTensor::singleton())
       pImpl->release();
   }
   TensorBase & operator=(TensorBase && rhs) & {
@@ -48,6 +51,9 @@ struct TensorBase {
   TensorImpl * get() const {
     return pImpl;
   }
+  bool defined() const {
+    return pImpl != UndefinedTensor::singleton();
+  }
 
   friend struct Type;
 

torch/lib/ATen/TensorOperators.h

@@ -11,6 +11,7 @@ inline Tensor & Tensor::operator=(Scalar v) && {
   return assign_(v);
 }
 inline Tensor & Tensor::assign_(Scalar v) {
+  AT_ASSERT(defined(), "attempting to assign a scalar to an undefined tensor");
   AT_ASSERT(dim() == 0, "attempting to assign a scalar to %d dim tensor", dim());
   pImpl->assign_(v);
   return *this;

torch/lib/ATen/UndefinedTensor.cpp

@@ -0,0 +1,42 @@
+#include "ATen/UndefinedTensor.h"
+#include "ATen/Context.h"
+
+namespace at {
+
+// should this use the globalContext?  Can it get a context passed in somehow?
+UndefinedTensor::UndefinedTensor()
+: TensorImpl(&(globalContext().getType(Backend::Undefined,ScalarType::Undefined))) {
+}
+
+const char * UndefinedTensor::toString() const {
+  return "UndefinedTensor";
+}
+
+IntList UndefinedTensor::sizes() const {
+  runtime_error("sizes() called on undefined Tensor");
+}
+
+int64_t UndefinedTensor::dim() const {
+  runtime_error("dim() called on undefined Tensor");
+}
+
+const char * UndefinedTensor::typeString() {
+  return "UndefinedType";
+}
+void * UndefinedTensor::unsafeGetTH(bool retain) {
+  runtime_error("unsafeGetTH(bool retain) called on undefined Tensor");
+}
+
+IntList UndefinedTensor::strides() const {
+  runtime_error("strides() called on undefined Tensor");
+}
+Scalar UndefinedTensor::localScalar() {
+  runtime_error("localScalar() called on undefined Tensor");
+}
+void UndefinedTensor::assign_(Scalar s) {
+  runtime_error("assign_() called on undefined Tensor");
+}
+
+UndefinedTensor UndefinedTensor::_singleton;
+
+}

torch/lib/ATen/UndefinedTensor.h

@@ -0,0 +1,28 @@
+#pragma once
+
+#include "ATen/TensorImpl.h"
+
+namespace at {
+
+struct UndefinedTensor final : public TensorImpl {
+public:
+  static inline UndefinedTensor * singleton() {
+    return &_singleton;
+  }
+  virtual ~UndefinedTensor() {}
+  virtual const char * toString() const override;
+  virtual IntList sizes() const override;
+  virtual IntList strides() const override;
+  virtual int64_t dim() const override;
+  virtual Scalar localScalar() override;
+  virtual void assign_(Scalar s) override;
+  virtual void * unsafeGetTH(bool retain) override;
+  static const char * typeString();
+private:
+  UndefinedTensor();
+  static UndefinedTensor _singleton;
+public:
+  friend struct UndefinedType;
+};
+
+} // namespace at

torch/lib/ATen/UndefinedType.cpp

@@ -0,0 +1,65 @@
+#include "ATen/UndefinedType.h"
+
+namespace at {
+
+UndefinedType::UndefinedType(Context* context)
+: Type(context) {}
+ScalarType UndefinedType::scalarType() const {
+  return ScalarType::Undefined;
+}
+Backend UndefinedType::backend() const {
+  return Backend::Undefined;
+}
+bool UndefinedType::isCuda() const { return false; }
+bool UndefinedType::isSparse() const { return false; }
+bool UndefinedType::isDistributed() const { return false; }
+
+std::unique_ptr<Storage> UndefinedType::storage() const {
+  runtime_error("storage not defined for UndefinedType");
+}
+std::unique_ptr<Storage> UndefinedType::storage(size_t size) const {
+  runtime_error("storage(size_t) not defined for UndefinedType");
+}
+std::unique_ptr<Storage> UndefinedType::storageFromBlob(void * data, int64_t size, const std::function<void(void*)> & deleter) const {
+  runtime_error("storageFromBlob not defined for UndefinedType");
+}
+Tensor UndefinedType::unsafeTensorFromTH(void * th_pointer, bool retain) const {
+  runtime_error("unsafeTensorFromTH not defined for UndefinedType");
+}
+std::unique_ptr<Generator> UndefinedType::generator() const {
+  runtime_error("generator not defined for UndefinedType");
+}
+
+const char * UndefinedType::toString() const {
+  return UndefinedType::typeString();
+}
+TypeID UndefinedType::ID() const {
+  return TypeID::Undefined;
+}
+
+std::size_t UndefinedType::elementSizeInBytes() const {
+  runtime_error("elementSizeInBytes not defined for UndefinedType");
+}
+
+Type & UndefinedType::toBackend(Backend b) const {
+  if (b == Backend::Undefined) {
+    return Type::toBackend(b);
+  }
+  runtime_error("toBackend not implemented for UndefinedType to non-UndefinedType");
+}
+Type & UndefinedType::toScalarType(ScalarType s) const {
+  if (s == ScalarType::Undefined) {
+    return Type::toScalarType(s);
+  }
+  runtime_error("toScalarType not implemented for UndefinedType to non-UndefinedType");
+}
+
+const char * UndefinedType::typeString() {
+  return "UndefinedType";
+}
+
+void UndefinedType::s_copy(const Tensor & src, Tensor & dst) const {
+  runtime_error("s_copy not defined for UndefinedType");
+}
+
+}

torch/lib/ATen/UndefinedType.h

@@ -0,0 +1,37 @@
+#pragma once
+
+#include "ATen/Type.h"
+#include "ATen/Context.h"
+#include "ATen/CheckGenerator.h"
+
+#ifdef _MSC_VER
+#ifdef Type
+#undef Type
+#endif
+#endif
+
+namespace at {
+
+struct UndefinedType final : public Type {
+  explicit UndefinedType(Context* context);
+  virtual ScalarType scalarType() const override;
+  virtual Backend backend() const override;
+  virtual bool isCuda() const override;
+  virtual bool isSparse() const override;
+  virtual bool isDistributed() const override;
+  virtual std::unique_ptr<Storage> storage() const override;
+  virtual std::unique_ptr<Storage> storage(size_t size) const override;
+  virtual std::unique_ptr<Storage> storageFromBlob(void * data, int64_t size, const std::function<void(void*)> & deleter) const override;
+  virtual std::unique_ptr<Generator> generator() const override;
+  virtual const char * toString() const override;
+  virtual std::size_t elementSizeInBytes() const override;
+  virtual Type & toBackend(Backend b) const;
+  virtual Type & toScalarType(ScalarType s) const;
+  virtual TypeID ID() const override;
+  static const char * typeString();
+  Tensor unsafeTensorFromTH(void * th_pointer, bool retain) const override;
+
+  virtual void s_copy(const Tensor & src, Tensor & dst) const override;
+};
+
+} // namespace at

torch/lib/ATen/Utils.h

@@ -2,6 +2,7 @@
 
 #include "ArrayRef.h"
 #include "ATenGeneral.h"
+#include "UndefinedTensor.h"
 #include <algorithm>
 #include <sstream>
 #include <typeinfo>
@@ -14,13 +15,17 @@ namespace at {
 AT_API void runtime_error(const char *format, ...);
 
 template <typename T, typename Base>
-static inline T* checked_cast(Base* expr, const char * name, int pos, bool allowNull) {
-  if(!expr) {
-    if (allowNull) {
-      return (T*) expr;
-    }
-    runtime_error("Expected a Tensor of type %s but found an undefined Tensor for argument #%d '%s'",
-      T::typeString(),pos,name);
+static inline T* checked_cast_storage(Base* expr, const char * name, int pos) {
+  if (typeid(*expr) != typeid(T))
+    runtime_error("Expected object of type %s but found type %s for argument #%d '%s'",
+      T::typeString(),expr->type().toString(),pos,name);
+  return static_cast<T*>(expr);
+}
+
+template <typename T, typename Base>
+inline T* checked_cast_tensor(Base* expr, const char * name, int pos, bool allowNull) {
+  if(allowNull && expr == UndefinedTensor::singleton()) {
+    return nullptr;
   }
   if (typeid(*expr) != typeid(T))
     runtime_error("Expected object of type %s but found type %s for argument #%d '%s'",
@@ -34,11 +39,6 @@ static inline std::vector<TH*> tensor_list_checked_cast(ArrayRef<TBase> tensors,
   std::vector<TH*> casted(tensors.size());
   for (unsigned int i = 0; i < tensors.size(); ++i) {
     auto *expr = tensors[i].pImpl;
-    if (!expr) {
-      runtime_error("Expected a Tensor of type %s but found an undefined Tensor for sequence element %u "
-                    " in sequence argument at position #%d '%s'",
-                    T::typeString(),i,pos,name);
-    }
     auto result = dynamic_cast<T*>(expr);
     if (result) {
       casted[i] = result->tensor;

torch/lib/ATen/copy_wrapper.py

@@ -25,7 +25,7 @@ case ${src_id}:
 FUNCTION = CodeTemplate("""\
 void ${Type}::s_copy(const Tensor & src, Tensor & dst) const {
   // code generated by function_wrapper
-  auto dst_ = checked_cast<${Tensor}>(dst.pImpl,"dst",0,false);
+  auto dst_ = checked_cast_tensor<${Tensor}>(dst.pImpl,"dst",0,false);
   (void) dst_; //silence unused warning
   switch(src.type().ID()) {
     ${copy_body}

torch/lib/ATen/function_wrapper.py

@@ -19,7 +19,7 @@ ${return_type} ${method_prefix}${api_name}(${formals_with_defaults}) const;
 TYPE_METHOD_DEFINITION_BROADCAST = CodeTemplate("""\
 ${return_type} Type::${method_prefix}${api_name}(${formals}) const {
     Tensor ${broadcast_returns};
-    std::tie(${broadcast_returns}) = ${broadcast_function}(${broadcast_actuals});
+    std::tie(${broadcast_returns}) = ${broadcast_function}(${broadcast_actuals}, "${api_name}");
     return ${method_prefix_derived}${api_name}(${broadcast_modified_actuals});
 }
 """)
@@ -142,20 +142,22 @@ TYPE_RETURN = {
 }
 
 CHECKED_CAST = {
-    'THTensor*': CodeTemplate('checked_cast<${Tensor}>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
+    'THTensor*':
+        CodeTemplate(
+            'checked_cast_tensor<${Tensor}>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
     'THSTensor*':
     CodeTemplate(
-        'checked_cast<Sparse${Tensor}>(${arg_name}.tref.pImpl,"${arg_name}",${arg_pos},false)'),
+        'checked_cast_tensor<Sparse${Tensor}>(${arg_name}.tref.pImpl,"${arg_name}",${arg_pos},false)'),
     'THBoolTensor*':
         CodeTemplate(
-            'checked_cast<${Backend}ByteTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
+            'checked_cast_tensor<${Backend}ByteTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
     'THIndexTensor*':
         CodeTemplate(
-            'checked_cast<${Backend}LongTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
+            'checked_cast_tensor<${Backend}LongTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
     'THIntegerTensor*':
         CodeTemplate(
-            'checked_cast<${Backend}IntTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
-    'THStorage*': CodeTemplate('checked_cast<${Storage}>(&${arg_name},"${arg_name}",${arg_pos}, false)'),
+            'checked_cast_tensor<${Backend}IntTensor>(${arg_name}.pImpl,"${arg_name}",${arg_pos}, ${null_okay})'),
+    'THStorage*': CodeTemplate('checked_cast_storage<${Storage}>(&${arg_name},"${arg_name}",${arg_pos})'),
     'THGenerator*':
         CodeTemplate(
             'check_generator<${Backend}Generator>(${arg_name}, &context->defaultGenerator(backend()))'),
@@ -720,11 +722,14 @@ def create_derived(backend_type_env, declarations):
     def allocate_arg(env, arg, output_count):
         name = arg['name']
         allocation = CodeTemplate(ALLOC_WRAP[arg['type']]).substitute(env)
+        tensor_arg = '{}_'.format(name)
         if arg.get('mask', False):
             allocation = 'output_mask[{}] ? {} : nullptr'.format(output_count, allocation)
+            tensor_arg = ('{}_ == nullptr ? (TensorImpl*)UndefinedTensor::singleton() : (TensorImpl*){}_'
+                          .format(name, name))
         return [
             'auto {}_ = {};'.format(name, allocation),
-            'auto {} = Tensor({}_,false);'.format(name, name),
+            'auto {} = Tensor({}, false);'.format(name, tensor_arg),
         ]
 
     def resize_arg(arg):

torch/lib/ATen/nn.yaml

@@ -3,7 +3,7 @@
 - name: binary_cross_entropy(Tensor input, Tensor target, Tensor weight={}, bool size_average=true)
   cname: BCECriterion
 
-- name: kl_div(Tensor input, Tensor target, bool size_average=true)
+- name: kl_div(Tensor input, Tensor target, bool size_average=true, bool reduce=true)
   cname: DistKLDivCriterion
 
 - name: l1_loss(Tensor input, Tensor target, bool size_average=true, bool reduce=True)
@@ -58,6 +58,8 @@
 
 - name: log_softmax(Tensor input, int64_t dim)
   cname: LogSoftMax
+  wrap_dim:
+    dim: input
 
 - name: prelu(Tensor input, Tensor weight)
   cname: PReLU
@@ -68,6 +70,8 @@
 
 - name: softmax(Tensor input, int64_t dim)
   cname: SoftMax
+  wrap_dim:
+    dim: input
 
 - name: softplus(Tensor input, Scalar beta=1, Scalar threshold=20)
   cname: SoftPlus

torch/lib/ATen/nn_parse.py

@@ -171,6 +171,8 @@ def get_thnn_args(thnn_function, params):
             thnn_args.append(arg_expr(name[0], name[1:]))
         elif name == 'scale':
             thnn_args.append({'type': 'EXPRESSION', 'name': '1'})
+        elif name == 'inplace':
+            thnn_args.append({'type': 'EXPRESSION', 'name': 'false'})
         else:
             raise RuntimeError("{}: can't find binding for '{}'"
                                .format(thnn_function.name, name))
@@ -261,7 +263,8 @@ def backward_declaration(base, thnn_functions):
 
     arguments = []
     arguments.append({'type': 'THTensor*', 'name': 'grad_output'})
-    arguments += [copy.deepcopy(arg) for arg in base['arguments']]
+    arguments += [copy.deepcopy(arg) for arg in base['arguments']
+                  if arg['name'] != 'inplace']
     arguments += base['buffers']
 
     for arg in arguments:

torch/lib/ATen/templates/Tensor.h

@@ -70,9 +70,6 @@ struct Tensor : public detail::TensorBase {
     pImpl = nullptr;
     return ret;
   }
-  bool defined() const {
-    return pImpl != nullptr;
-  }
   void swap(Tensor & rhs) {
     TensorImpl * tmp = pImpl;
     pImpl = rhs.pImpl;

torch/lib/ATen/templates/Type.cpp

@@ -5,6 +5,7 @@
 #include "ATen/SparseTensorRef.h"
 #include "ATen/ExpandUtils.h"
 #include "ATen/NativeFunctions.h"
+#include "ATen/UndefinedType.h"
 
 #include <iostream>
 ${type_headers}
@@ -13,15 +14,17 @@ namespace at {
 
 void Type::registerAll(Context * context) {
   ${type_registrations}
+  context->type_registry[static_cast<int>(Backend::Undefined)][static_cast<int>(ScalarType::Undefined)].reset(new UndefinedType(context));
 }
 
 void Type::copy(const Tensor & src, Tensor & dst) const {
   Tensor b_src;
-  std::tie(b_src) = expand_inplace(dst, src);
+  std::tie(b_src) = expand_inplace(dst, src, "copy");
   s_copy(b_src, dst);
 }
 
 Tensor Type::copy(const Tensor & src) const {
+  AT_ASSERT(src.defined(), "attempt to copy an undefined tensor");
   Tensor r = this->tensor(src.sizes());
   r.copy_(src);
   return r;

torch/lib/ATen/templates/Type.h

@@ -56,6 +56,7 @@ static inline void noop_deleter(void*) {}
 
 enum class TypeID {
   ${type_ids}
+  Undefined,
   NumOptions
 };
 

torch/lib/ATen/templates/TypeDerived.cpp

@@ -9,6 +9,7 @@
 #include "ATen/Utils.h"
 #include "ATen/WrapDimUtils.h"
 #include "ATen/THLongStorageView.h"
+#include "ATen/UndefinedTensor.h"
 #include <iostream>
 #include <sstream>
 

torch/lib/ATen/test/CMakeLists.txt

@@ -18,3 +18,6 @@ target_link_libraries(dlconvertor_test ATen)
 
 add_executable(native_test native_test.cpp)
 target_link_libraries(native_test ATen)
+
+add_executable(undefined_tensor_test undefined_tensor_test.cpp)
+target_link_libraries(undefined_tensor_test ATen)

torch/lib/ATen/test/undefined_tensor_test.cpp

@@ -0,0 +1,60 @@
+#include "ATen/ATen.h"
+#include "ATen/UndefinedTensor.h"
+#include <string>
+#include "test_assert.h"
+
+
+using namespace at;
+
+#define ASSERT_THROWS(fn, message)                                  \
+try {                                                               \
+  fn;                                                               \
+  ASSERT(false);                                                    \
+} catch(std::runtime_error &e) {                                    \
+  ASSERT(std::string(e.what()).find(message) != std::string::npos); \
+}
+
+
+int main() {
+  // mainly test ops on undefined tensors don't segfault and give a reasonable errror message.
+  Tensor und;
+  Tensor ft = CPU(kFloat).ones({1});
+
+  std::cout << und << std::endl;
+  ASSERT(!und.defined());
+  ASSERT(std::string("UndefinedTensor") == und.toString());
+
+  ASSERT_THROWS(und.strides(), "strides");
+  ASSERT_THROWS(und.dim(), "dim");
+  ASSERT_THROWS(und.assign_(Scalar(5)), "assign");
+  ASSERT_THROWS(und.unsafeGetTH(true), "unsafeGetTH");
+  ASSERT_THROWS(und.add(und), "add");
+  ASSERT_THROWS(und.add(ft), "add");
+  ASSERT_THROWS(ft.add(und), "add");
+  ASSERT_THROWS(und.add(5), "add");
+  ASSERT_THROWS(und.mm(und), "mm");
+
+  und.toType(und.type());
+  ASSERT_THROWS(und.toType(ft.type()), "attempt to copy an undefined tensor");
+  ASSERT_THROWS(ft.toType(und.type()), "UndefinedType");
+  und.toType(ScalarType::Undefined);
+  ASSERT_THROWS(und.toType(ScalarType::Float), "toScalarType");
+  ASSERT_THROWS(ft.toType(ScalarType::Undefined), "UndefinedType");
+
+  // copy_
+  ASSERT_THROWS(und.copy_(und), "copy");
+  ASSERT_THROWS(und.copy_(ft), "copy");
+  ASSERT_THROWS(ft.copy_(und), "copy");
+
+  und.toBackend(Backend::Undefined);
+  ASSERT_THROWS(und.toBackend(Backend::CPU), "toBackend");
+  ASSERT_THROWS(ft.toBackend(Backend::Undefined), "UndefinedType");
+
+  Tensor to_move = CPU(kFloat).ones({1});
+  Tensor m(std::move(to_move));
+  ASSERT(!to_move.defined());
+  ASSERT(to_move.get() == UndefinedTensor::singleton());
+
+  return 0;
+}
+

torch/lib/TH/CMakeLists.txt

@@ -306,6 +306,9 @@ IF(BLAS_FOUND)
   IF ($ENV{TH_BINARY_BUILD})
     MESSAGE(STATUS "TH_BINARY_BUILD detected. Enabling special linkage.")
     TARGET_LINK_LIBRARIES(TH "${BLAS_LIBRARIES};${BLAS_LIBRARIES};${BLAS_LIBRARIES}")
+    IF (UNIX AND NOT APPLE)
+      set (CMAKE_SHARED_LINKER_FLAGS "-Wl,--version-script=${CMAKE_CURRENT_SOURCE_DIR}/../../../tools/pytorch.version")
+    ENDIF(UNIX AND NOT APPLE)
   ELSE ($ENV{TH_BINARY_BUILD})
     TARGET_LINK_LIBRARIES(TH ${BLAS_LIBRARIES})
   ENDIF ($ENV{TH_BINARY_BUILD})

torch/lib/TH/THDiskFile.c

@@ -2,6 +2,10 @@
 #include "THDiskFile.h"
 #include "THFilePrivate.h"
 
+#ifndef _WIN32
+#include <sys/types.h>
+#endif
+
 #include <stdint.h>
 #ifndef LLONG_MAX
 #define LLONG_MAX 9223372036854775807LL

torch/lib/TH/THMemoryFile.c

@@ -2,6 +2,10 @@
 #include "THFilePrivate.h"
 #include "stdint.h"
 
+#ifndef _WIN32
+#include <sys/types.h>
+#endif
+
 typedef struct THMemoryFile__
 {
     THFile file;