Add copyright header and pre-commit validator (#1900)

Author: jainapurva

.pre-commit-config.yaml

@@ -28,3 +28,12 @@ repos:
           - --isolated
           - --select
           - F821,F823,W191
+
+  - repo: local
+    hooks:
+      - id: privacy-policy-check
+        name: Check Privacy Policy Headers
+        entry: python scripts/check_copyright_header.py
+        language: python
+        types_or: [python, c, c++, shell, text]
+        files: \.(py|cu|h|cuh|sh|metal)$

benchmarks/bench_galore_fused_kernels.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import argparse
 import os
 

benchmarks/benchmark_aq.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 """Benchmarks for affine quantized tensor, this includes int8 dynamic quant, int8 weight only quant and int4 weight only quant APIs"""
 
 import copy

benchmarks/benchmark_fp6.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import pandas as pd
 import torch
 import torch.nn.functional as F

benchmarks/benchmark_gpu_sparsity.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import argparse
 
 import pandas as pd

benchmarks/benchmark_hqq.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 try:
     import hqq  # noqa: F401
     import triton

benchmarks/benchmark_low_bit_adam.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 # pip install timm wandb tqdm datasets bitsandbytes
 #
 # optional:

benchmarks/benchmark_marlin_qqq.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import pandas as pd
 import torch
 from tqdm import tqdm

benchmarks/benchmark_rowwise_scaled_linear_cutlass.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import pandas as pd
 import torch
 from tqdm import tqdm

benchmarks/benchmark_rowwise_scaled_linear_sparse_cutlass.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import pandas as pd
 import torch
 from tqdm import tqdm

benchmarks/benchmark_uintx.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 from copy import deepcopy
 
 import torch
@@ -62,16 +67,16 @@ def profile_bitpack():
     prof.export_chrome_trace("trace.json")
     """
     CPU perf:
-        unpack_gpu 
+        unpack_gpu
         Self CPU time total: 602.501ms
-        
-        unpack_cpu 
+
+        unpack_cpu
         Self CPU time total: 415.469ms
     GPU perf:
-        unpack_gpu  on gpu: 
+        unpack_gpu  on gpu:
         Self CPU time total: 58.512ms
         Self CUDA time total: 5.083ms
-    
+
         unpack_cpu:
         Self CPU time total: 96.947ms
         Self CUDA time total: 5.253ms

benchmarks/float8/bench_padding.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 from dataclasses import dataclass
 from typing import Optional
 

benchmarks/float8/training/float8_training_benchmark.sh

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 #!/bin/bash
 # This script can be used to launch a torchtitan float8 training run
 # with the given parameters,

benchmarks/float8/training/parse_torchtitan_logs.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 #!/usr/bin/env python3
 """
 Script which can be used to parse the log file generated by the torchtitan,

benchmarks/fused_benchmark_utils.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import torch
 import triton
 from triton.testing import do_bench

benchmarks/intmm.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import argparse
 import csv
 import itertools

benchmarks/print_config_shapes.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 from torchao.kernel import autotuner
 
 configs = autotuner._load_best_configs()

benchmarks/quantized_training/benchmark_int8mm.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import pandas as pd
 import torch
 from triton.testing import do_bench

benchmarks/quantized_training/pretrain_llama2.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 # pre-train a mini Llama2 on TinyStories with INT8 quantized training
 # pip install huggingface_hub sentencepiece wandb
 #

docs/source/tutorials_source/template_tutorial.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 # -*- coding: utf-8 -*-
 
 """

examples/sam2_amg_server/amg_example.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import cv2
 import matplotlib.pyplot as plt
 import numpy as np

examples/sam2_amg_server/annotate_with_rle.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import json
 from datetime import datetime
 from io import BytesIO

examples/sam2_amg_server/cli.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 from io import BytesIO
 
 import fire

examples/sam2_amg_server/cli_on_modal.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import asyncio
 import json
 from pathlib import Path

examples/sam2_amg_server/compare_rle_lists.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import json
 from pathlib import Path
 from typing import Any, Dict

examples/sam2_amg_server/compile_export_utils.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import time
 from pathlib import Path
 from typing import Optional

examples/sam2_amg_server/generate_data.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import json
 import time
 from collections import OrderedDict

examples/sam2_amg_server/modal_experiments.sh

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 #!/bin/bash
 
 set -ex
@@ -40,7 +45,7 @@ modal app stop torchao-sam-2-cli-mps
 
 echo "amg vs baseline"
 python compare_rle_lists.py ~/blogs/outputs/amg  ~/blogs/outputs/amg_baseline --compare-folders --strict
-echo "sps vs baseline"                                               
+echo "sps vs baseline"
 python compare_rle_lists.py ~/blogs/outputs/sps  ~/blogs/outputs/sps_baseline --compare-folders --strict
-echo "mps vs baseline"                                               
+echo "mps vs baseline"
 python compare_rle_lists.py ~/blogs/outputs/mps  ~/blogs/outputs/mps_baseline --compare-folders --strict

examples/sam2_amg_server/reproduce_experiments.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import json
 import os
 import subprocess

examples/sam2_amg_server/server.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import asyncio
 import json
 import logging

examples/sam2_vos_example/compile_export_utils.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import time
 from pathlib import Path
 from typing import Optional

examples/sam2_vos_example/video_profile.py

@@ -1,3 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD 3-Clause license found in the
+# LICENSE file in the root directory of this source tree.
 import os
 import time
 from datetime import datetime