[sync] Issue template + Pyre

.github/ISSUE_TEMPLATE/bug-report.md

@@ -0,0 +1,53 @@
+---
+name: "\U0001F41B Bug Report"
+about: Submit a bug report to help us improve xFormers
+
+---
+
+# 🐛 Bug
+
+<!-- A clear and concise description of what the bug is. -->
+
+## Command
+
+## To Reproduce
+
+Steps to reproduce the behavior:
+
+<!-- If you were running a command, post the exact command that you were running -->
+
+1.
+2.
+3.
+
+<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
+
+## Expected behavior
+
+<!-- A clear and concise description of what you expected to happen. -->
+
+## Environment
+
+Please copy and paste the output from the
+environment collection script from PyTorch
+(or fill out the checklist below manually).
+
+You can run the script with:
+
+```bash
+# For security purposes, please check the contents of collect_env.py before running it.
+python -m torch.utils.collect_env
+```
+
+- PyTorch Version (e.g., 1.0):
+- OS (e.g., Linux):
+- How you installed PyTorch (`conda`, `pip`, source):
+- Build command you used (if compiling from source):
+- Python version:
+- CUDA/cuDNN version:
+- GPU models and configuration:
+- Any other relevant information:
+
+## Additional context
+
+<!-- Add any other context about the problem here. -->

.github/ISSUE_TEMPLATE/feature-request.md

@@ -0,0 +1,25 @@
+---
+name: "\U0001F680Feature Request"
+about: Submit a proposal/request for a new xFormers feature
+
+---
+
+# 🚀 Feature
+
+<!-- A clear and concise description of the feature proposal -->
+
+## Motivation
+
+<!-- Please outline the motivation for the proposal. Is your feature request related to a problem? e.g., I'm always frustrated when [...]. If this is related to another GitHub issue, please link here too -->
+
+## Pitch
+
+<!-- A clear and concise description of what you want to happen. -->
+
+## Alternatives
+
+<!-- A clear and concise description of any alternative solutions or features you've considered, if any. -->
+
+## Additional context
+
+<!-- Add any other context or screenshots about the feature request here. -->

.github/ISSUE_TEMPLATE/questions-help-support.md

@@ -0,0 +1,7 @@
+---
+name: "❓Questions/Help/Support"
+about: Do you need support?
+
+---
+
+# ❓ Questions and Help

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,21 @@
+## What does this PR do?
+Fixes # (issue).
+
+## Before submitting
+
+- [ ] Did you have fun?
+  - Make sure you had fun coding 🙃
+- [ ] Did you read the [contributor guideline](https://github.com/facebookresearch/fairscale/blob/master/CONTRIBUTING.md)?
+- [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements)
+  - [ ] N/A
+- [ ] Did you make sure to update the docs?
+  - [ ] N/A
+- [ ] Did you write any new necessary tests?
+  - [ ] N/A
+- [ ] Did you update the [changelog](https://github.com/facebookresearch/fairscale/blob/master/CHANGELOG.md)? (if needed)
+  - [ ] N/A
+
+
+## PR review
+Anyone in the community is free to review the PR once the tests have passed.
+If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

.github/workflows/gh-pages.yml

@@ -49,5 +49,5 @@ jobs:
         uses: peaceiris/actions-gh-pages@v3
         with:
           github_token: ${{ secrets.GITHUB_TOKEN }}
-          publish_dir: ./public
+          publish_dir: docs/build/html
         if: github.event_name != 'pull_request'

.gitignore

@@ -38,3 +38,5 @@ my_runs.md
 # JetBrains PyCharm IDE
 .idea/
 
+# Pyre cache.
+.pyre/

.pyre_configuration

@@ -0,0 +1,7 @@
+{
+  "python_version": "3.7",
+  "source_directories": [
+    "stubs",
+    "."
+  ]
+}

.watchmanconfig

@@ -0,0 +1 @@
+{}

CODE_OF_CONDUCT.md

@@ -0,0 +1,80 @@
+# Code of Conduct
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as
+contributors and maintainers pledge to make participation in our project and
+our community a harassment-free experience for everyone, regardless of age, body
+size, disability, ethnicity, sex characteristics, gender identity and expression,
+level of experience, education, socio-economic status, nationality, personal
+appearance, race, religion, or sexual identity and orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment
+include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or
+advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic
+address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a
+professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable
+behavior and are expected to take appropriate and fair corrective action in
+response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or
+reject comments, commits, code, wiki edits, issues, and other contributions
+that are not aligned to this Code of Conduct, or to ban temporarily or
+permanently any contributor for other behaviors that they deem inappropriate,
+threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies within all project spaces, and it also applies when
+an individual is representing the project or its community in public spaces.
+Examples of representing a project or community include using an official
+project e-mail address, posting via an official social media account, or acting
+as an appointed representative at an online or offline event. Representation of
+a project may be further defined and clarified by project maintainers.
+
+This Code of Conduct also applies outside the project spaces when there is a
+reasonable belief that an individual's behavior may have a negative impact on
+the project or its community.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported by contacting the project team at <opensource-conduct@fb.com>. All
+complaints will be reviewed and investigated and will result in a response that
+is deemed necessary and appropriate to the circumstances. The project team is
+obligated to maintain confidentiality with regard to the reporter of an incident.
+Further details of specific enforcement policies may be posted separately.
+
+Project maintainers who do not follow or enforce the Code of Conduct in good
+faith may face temporary or permanent repercussions as determined by other
+members of the project's leadership.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
+available at https://www.contributor-covenant.org/version/1/4/code-of-conduct.html
+
+[homepage]: https://www.contributor-covenant.org
+
+For answers to common questions about this code of conduct, see
+https://www.contributor-covenant.org/faq

docs/requirements.txt

@@ -4,3 +4,4 @@ sphinx_rtd_theme==0.4.3
 sphinxcontrib-programoutput==0.16
 git+https://github.com/pytorch/pytorch_sphinx_theme.git#egg=pytorch_sphinx_theme
 torch>=1.6.0
+numpy>=1.19.5

docs/source/tutorials/pytorch_encoder.rst

@@ -74,50 +74,7 @@ You can think of it as a declaration of the sequence of blocks that you would li
                     "attention": {
                         "name": "linformer", # whatever attention mechanism
                         "dropout": 0,
-                        "causal": True,
-                        "seq_len": 512,
-                    },
-                },
-                "feedforward_config": {
-                    "name": "MLP",
-                    "dim_model": 384,
-                    "dropout": 0,
-                    "activation": "relu",
-                    "hidden_layer_multiplier": 4,
-                },
-            }
-        }
-    ]
-
-    config = xFormerConfig(**my_config)  # This part of xFormers is entirely type checked and needs a config object, could be changed in the fututure
-    model = xFormer.from_config(config).to(device)
-
-    from xformers.factory.model_factory import xFormer, xFormerConfig
-
-    my_config =  [
-        # A list of the encoder or decoder blocks which constitute the Transformer.
-        # Note that a sequence of different encoder blocks can be used, same for decoders
-        {
-            "reversible": False,  # Optionally make these layers reversible, to save memory
-            "block_config": {
-                "block_type": "encoder",
-                "num_layers": 3,  # Optional, this means that this config will repeat N times
-                "dim_model": 384,
-                "layer_norm_style": "pre",  # Optional, pre/post
-                "position_encoding_config": {
-                    "name": "vocab",  # whatever position encodinhg makes sense
-                    "dim_model": 384,
-                    "seq_len": 1024,
-                    "vocab_size": 64,
-                },
-                "multi_head_config": {
-                    "num_heads": 4,
-                    "dim_model": 384,
-                    "residual_dropout": 0,
-                    "attention": {
-                        "name": "linformer", # whatever attention mechanism
-                        "dropout": 0,
-                        "causal": True,
+                        "causal": False,
                         "seq_len": 512,
                     },
                 },

requirements-test.txt

@@ -7,6 +7,7 @@ black == 20.8b1
 flake8 == 3.8.4
 isort == 5.7.0
 mypy == 0.812
+pyre-check == 0.9.6
 
 # Tools for unit tests & coverage.
 pytest == 5.4.1

stubs/fvcore/nn.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/matplotlib/pyplot.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/pandas.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/recommonmark/transform.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/seaborn.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/sklearn/model_selection.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/submitit.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/tensorflow.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/tqdm.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/triton/__init__.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/triton/language.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

stubs/triton/ops/blocksparse.pyi

@@ -0,0 +1,3 @@
+from typing import Any
+
+def __getattr__(name) -> Any: ...

tests/test_attentions.py

@@ -154,6 +154,7 @@ def test_different_kq_dimensions(
         "linformer",
         "blocksparse",
     }:
+        # pyre-fixme[29]: The library function `pytest.skip` is not supported by Pyre.
         pytest.skip(f"{attention_name} does not support different k, q dimensions yet.")
     multi_head = _get_multihead(attention_name, 0.0, 0.0, False, heads, device)
 

tests/test_feedforward.py

@@ -40,6 +40,7 @@ def test_feedforward(
     ffw = build_feedforward(test_config)
 
     if ffw.requires_cuda and not device.type == "cuda":
+        # pyre-fixme[29]: The library function `pytest.skip` is not supported by Pyre.
         pytest.skip("This MLP requires CUDA and current device does not match")
 
     inputs = torch.rand(BATCH, SEQ, LATENT, device=device)

tests/test_triton_fused_linear.py

@@ -98,6 +98,8 @@ def test_fused_linear_parity(shape, activation: Activation, bias: bool, amp: boo
     X_ = torch.normal(0, 1, size=shape, device="cuda")
     X_.requires_grad_()
 
+    # pyre-ignore[16]: TODO(T101400990): Pyre did not recognize the
+    # `FusedLinear` import.
     triton_fused_linear = FusedLinear(
         shape[-1], shape[-1] // 2, bias=bias, activation=activation
     ).to("cuda")

xformers/benchmarks/benchmark_triton_blocksparse.py

@@ -64,6 +64,8 @@ def bench_matmul(dtype: torch.dtype, shapes):
             if mode == "sdd":
                 b_cs = b_cs.transpose(-2, -1)
 
+            # pyre-fixme[16]: TODO(T101400990): Pyre did not recognize the
+            # `SparseCS` import.
             sparse_cs_mask = SparseCS(
                 mask.flatten(start_dim=0, end_dim=1).contiguous(),
                 device=torch.device("cuda"),

xformers/benchmarks/benchmark_triton_layernorm.py

@@ -36,6 +36,8 @@ def bench_layernorm(backward: bool):
             # Pytorch layer norn
             torch_layernorm = torch.nn.LayerNorm([K]).to(dtype=dtype, device=device)
 
+            # pyre-ignore[16]: TODO(T101400990): Pyre did not recognize the
+            # `FusedLinearNorm` import.
             # Fused layernorm equivalent
             fused_layernorm = FusedLayerNorm([K]).to(dtype=dtype, device=device)