Not requiring all functions to have types allows a lot of 'Any' types to slip in - which poison types and make mypy unable to properly typecheck the code. I want to flip the default so that new files are required to have fully typed defs and we can have a burndown list of files that fail to require full types.
The preceding stack of PRs (cut up simply to limit the number of file changes per PR "reasonable") adds `# mypy: allow-untyped-defs` to any file which didn't immediately pass mypy with the flag flipped. Due to changing files and merge conflicts it will probably be necessary to have several passes through before landing this final PR which turns the option on.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127836
Approved by: https://github.com/oulgen, https://github.com/Skylion007
Summary: We need an implementation of RedisRemoteCacheBackend with the same API that we're using for FbMemcacheRemoteFxGraphCacheBackend. So we'll stop using the Triton implementation and adapt a version for use by inductor. I also renamed parameters and cache entries to match our cache terminology.
Test Plan: Ran this command twice and inspected log output to ensure I got cache hits:
```
TORCH_LOGS=+torch._inductor.codecache TORCHINDUCTOR_FX_GRAPH_REMOTE_CACHE=1 python benchmarks/dynamo/torchbench.py --performance --inductor --device cuda --training --amp --print-compilation-time --only dcgan
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127480
Approved by: https://github.com/oulgen
Type checking Python is a pain. Here are my learnings:
* The types for heavily polymorphic code is going to be verbose, no way around it. I originally was hoping I could lean on polymorphism with a bounded TypeVar to compactly write signatures for many of the ValueRanges methods, but I ran into some unworkaroundable mypy bugs. Writing out all the types explicitly and using `@overload` liberally works pretty well, so I think I recommend people do that instead of trying to do fancy things.
* Sympy is missing annotations for assumptions, because they are all metaprogrammed. I don't really relish maintaining a typeshed for sympy, so I wrote a small mypy plugin to add them in.
* GADT style refinement is... just not a good idea in practice. Mypy easily gets confused whether or not a return value from a refined section is allowed for the outer return type. So many of these have been replaced with less informative implementation types and more informative external types via overloads. Hopefully this is good for use sites.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118870
Approved by: https://github.com/Skylion007, https://github.com/albanD
Fixes https://github.com/pytorch/pytorch/issues/118129
Suppressions automatically added with
```
import re
with open("error_file.txt", "r") as f:
errors = f.readlines()
error_lines = {}
for error in errors:
match = re.match(r"(.*):(\d+):\d+: error:.*\[(.*)\]", error)
if match:
file_path, line_number, error_type = match.groups()
if file_path not in error_lines:
error_lines[file_path] = {}
error_lines[file_path][int(line_number)] = error_type
for file_path, lines in error_lines.items():
with open(file_path, "r") as f:
code = f.readlines()
for line_number, error_type in sorted(lines.items(), key=lambda x: x[0], reverse=True):
code[line_number - 1] = code[line_number - 1].rstrip() + f" # type: ignore[{error_type}]\n"
with open(file_path, "w") as f:
f.writelines(code)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Co-authored-by: Catherine Lee <csl@fb.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118533
Approved by: https://github.com/Skylion007, https://github.com/zou3519
Fixes https://github.com/pytorch/pytorch/issues/118129
Suppressions automatically added with
```
import re
with open("error_file.txt", "r") as f:
errors = f.readlines()
error_lines = {}
for error in errors:
match = re.match(r"(.*):(\d+):\d+: error:.*\[(.*)\]", error)
if match:
file_path, line_number, error_type = match.groups()
if file_path not in error_lines:
error_lines[file_path] = {}
error_lines[file_path][int(line_number)] = error_type
for file_path, lines in error_lines.items():
with open(file_path, "r") as f:
code = f.readlines()
for line_number, error_type in sorted(lines.items(), key=lambda x: x[0], reverse=True):
code[line_number - 1] = code[line_number - 1].rstrip() + f" # type: ignore[{error_type}]\n"
with open(file_path, "w") as f:
f.writelines(code)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118533
Approved by: https://github.com/Skylion007, https://github.com/zou3519
dmypy silently ignores follow_imports = skip, so to get parity between
dmypy and mypy we have to suck it up and type: ignore all of the sympy
typing problems.
The suppressions were added automatically with the following script generated by GPT-4:
```
import re
# Read the error file
with open("error_file.txt", "r") as f:
errors = f.readlines()
# Parse the lines with errors and error types
error_lines = {}
for error in errors:
match = re.match(r"(.*):(\d+):\d+: error:.*\[(.*)\]", error)
if match:
file_path, line_number, error_type = match.groups()
if file_path not in error_lines:
error_lines[file_path] = {}
error_lines[file_path][int(line_number)] = error_type
# Insert ignore comments in the source files
for file_path, lines in error_lines.items():
with open(file_path, "r") as f:
code = f.readlines()
for line_number, error_type in sorted(lines.items(), key=lambda x: x[0], reverse=True):
code[line_number - 1] = code[line_number - 1].rstrip() + f" # type: ignore[{error_type}]\n"
with open(file_path, "w") as f:
f.writelines(code)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118469
Approved by: https://github.com/Skylion007
ghstack dependencies: #118414, #118418, #118432, #118467, #118468
The original motivation for MYPYINDUCTOR was a faster type checking configuration that only checked a subset of files. With the removal of `follow_imports = ignore`, we are now able to use dmypy to do fast incremental typechecking, eliminating the need for this.
Perhaps erroneously, when I tee'ed up this PR I elected to delete the `follow_imports = skip` designations in the mypy-inductor.ini. This lead to a number of extra type error suppressions that I manually edited. You will need to review.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118432
Approved by: https://github.com/Skylion007
ghstack dependencies: #118414, #118418
This is a lot of files changed! Don't panic! Here's how it works:
* Previously, we set `follow_imports = silent` for our mypy.ini configuration. Per https://mypy.readthedocs.io/en/stable/running_mypy.html#follow-imports, what this does is whenever we have an import to a module which is not listed as a file to be typechecked in mypy, we typecheck it as normal but suppress all errors that occurred in that file.
* When mypy is run inside lintrunner, the list of files is precisely the files covered by the glob in lintrunner.toml, but with files in excludes excluded.
* The top-level directive `# mypy: ignore-errors` instructs mypy to typecheck the file as normal, but ignore all errors.
* Therefore, it should be equivalent to set `follow_imports = normal`, if we put `# mypy: ignore-errors` on all files that were previously excluded from the file list.
* Having done this, we can remove the exclude list from .lintrunner.toml, since excluding a file from typechecking is baked into the files themselves.
* torch/_dynamo and torch/_inductor were previously in the exclude list, because they were covered by MYPYINDUCTOR. It is not OK to mark these as `# mypy: ignore-errors` as this will impede typechecking on the alternate configuration. So they are temporarily being checked twice, but I am suppressing the errors in these files as the configurations are not quite the same. I plan to unify the configurations so this is only a temporary state.
* There were some straggler type errors after these changes somehow, so I fixed them as needed. There weren't that many.
In the future, to start type checking a file, just remove the ignore-errors directive from the top of the file.
The codemod was done with this script authored by GPT-4:
```
import glob
exclude_patterns = [
...
]
for pattern in exclude_patterns:
for filepath in glob.glob(pattern, recursive=True):
if filepath.endswith('.py'):
with open(filepath, 'r+') as f:
content = f.read()
f.seek(0, 0)
f.write('# mypy: ignore-errors\n\n' + content)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118414
Approved by: https://github.com/thiagocrepaldi, https://github.com/albanD
Since MYPYNOFOLLOW is about to turn on import following, there's no
reason to keep test_utils.py in the MYPYNOFOLLOW config. Moreover, I'm
not sure it still takes 10 minutes to typecheck this file; adding it to
the MYPY config takes `lintrunner --take MYPY --all-files` from 53s to
57s on my machine, which is substantial but not horrible. I guess we'll
see how it fares on CI.
(Note that we cannot simply merge MYPY and MYPYNOFOLLOW because the
latter config turns on `disallow_any_generics` and so is in that sense
stricter than the MYPY config.)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113745
Approved by: https://github.com/clee2000
This added the numpy typing plugin to mypy config so that we could
use it for DeviceMesh typing annotations
Please see https://github.com/pytorch/pytorch/pull/92931 about why we need this. For example, we are currently saving the DeviceMesh's mesh field as torch.Tensor, where when we do sth like:
```python
with FakeTensorMode():
device_mesh = DeviceMesh("cuda", torch.arange(4))
```
It would throw error because FakeTensorMode or any TorchDispatchMode tracks every tensor creation and interactions. While DeviceMesh just want to save a nd-array to record the mesh topology, and would like to avoid the interaction with subsystems like FakeTensor, so we want to support saving `mesh` as numpy array instead.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/92930
Approved by: https://github.com/ezyang, https://github.com/malfet
As per title.
Additionally we also introduce support for:
- Rectangular block sizes which are powers of 2 and at least 16 (triton's `dot` limitation).
- Batch support with broadcasting for either of the arguments.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88078
Approved by: https://github.com/cpuhrsch
As per title.
Additionally we also introduce support for:
- Rectangular block sizes which are powers of 2 and at least 16 (triton's `dot` limitation).
- Batch support with broadcasting for either of the arguments.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88078
Approved by: https://github.com/cpuhrsch
As per title.
Additionally we also introduce support for:
- Rectangular block sizes which are powers of 2 and at least 16 (triton's `dot` limitation).
- Batch support with broadcasting for either of the arguments.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88078
Approved by: https://github.com/cpuhrsch
### Summary:
This PR updates the design of APoT Observer, Quantizer, and Tensor to be more consistent with their uniform counterparts in the PyTorch framework. APoT Observer now calculates alpha as the max between the absolute values of the max and min values in the input tensor. APoT Quantizer is modified so its instance methods quantize_APoT and dequantize_APoT are called by their global method counterparts. APoT Tensor is modified to account for the new method definition of the `quantize_APoT` from APoT Quantizer.
### Test Plan:
Run APoT Observer class unit tests with: `python pytorch/test/quantization/core/experimental/test_nonuniform_observer.py`
Run APoT Quantize class unit tests with: `python pytorch/test/quantization/core/experimental/test_quantizer.py`
Run APoT Tensor class unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/80075
Approved by: https://github.com/jerryzh168
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74833
FSDP has 74 type ignores and more being added, it may be better to
exclude from typechecking until we can type it properly.
ghstack-source-id: 152354450
Test Plan: CI
Reviewed By: zhaojuanmao
Differential Revision: D35186441
fbshipit-source-id: d06d90a06dcdbd38033adc70d0e9374333d4dc9a
(cherry picked from commit 4b410e1480abeeb672ce14274215ef55bcf3285d)
