Add files for finetuning LLMs on CPUs blog

Author: Evaan2001

finetuning_on_cpu/.gitignore

@@ -0,0 +1,8 @@
+.DS_Store # only for macs
+bazel-finetune-repo
+bazel-out
+bazel-bin
+bazel-testlogs
+.bazelrc
+.env
+MODULE.bazel.lock

finetuning_on_cpu/BUILD

@@ -0,0 +1,4 @@
+exports_files(
+    ["python_provider.sh"],
+    visibility = ["//visibility:public"],  # Makes it visible to all packages
+)

finetuning_on_cpu/MODULE.bazel

@@ -0,0 +1,16 @@
+bazel_dep(name = "rules_python", version = "1.2.0")
+bazel_dep(name = "rules_shell", version = "0.4.1")
+
+python = use_extension("@rules_python//python/extensions:python.bzl", "python")
+python.toolchain(
+    python_version = "3.13",
+)
+
+pip = use_extension("@rules_python//python/extensions:pip.bzl", "pip")
+pip.parse(
+    hub_name = "pypi",
+    python_version = "3.13",
+    requirements_lock = "//:bazel_requirements_lock.txt",
+)
+
+use_repo(pip, "pypi")

finetuning_on_cpu/bazel_requirements_lock.txt

@@ -0,0 +1,157 @@
+
+aiohappyeyeballs==2.6.1
+    # via aiohttp
+aiohttp==3.11.14
+    # via
+    #   datasets
+    #   fsspec
+aiosignal==1.3.2
+    # via aiohttp
+async-timeout==5.0.1
+    # via aiohttp
+attrs==25.3.0
+    # via aiohttp
+certifi==2025.1.31
+    # via requests
+charset-normalizer==3.4.1
+    # via requests
+datasets==3.4.1
+    # via -r requirements.txt
+dill==0.3.7
+    # via
+    #   datasets
+    #   multiprocess
+filelock==3.18.0
+    # via
+    #   huggingface-hub
+    #   torch
+    #   transformers
+frozenlist==1.5.0
+    # via
+    #   aiohttp
+    #   aiosignal
+fsspec[http]==2024.12.0
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   torch
+huggingface-hub==0.29.3
+    # via
+    #   datasets
+    #   tokenizers
+    #   transformers
+idna==3.10
+    # via
+    #   requests
+    #   yarl
+jinja2==3.1.6
+    # via torch
+markupsafe==3.0.2
+    # via jinja2
+mpmath==1.3.0
+    # via sympy
+multidict==6.2.0
+    # via
+    #   aiohttp
+    #   yarl
+multiprocess==0.70.15
+    # via datasets
+networkx==3.2.1
+    # via torch
+numpy==1.26.4
+    # via
+    #   datasets
+    #   pandas
+    #   transformers
+packaging==24.2
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   transformers
+pandas==2.2.3
+    # via datasets
+propcache==0.3.0
+    # via
+    #   aiohttp
+    #   yarl
+pyarrow==19.0.1
+    # via datasets
+python-dateutil==2.9.0.post0
+    # via pandas
+pytz==2025.2
+    # via pandas
+pyyaml==6.0.2
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   transformers
+regex==2024.11.6
+    # via transformers
+requests==2.32.3
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   transformers
+safetensors==0.5.3
+    # via transformers
+six==1.17.0
+    # via python-dateutil
+sympy==1.13.1
+    # via torch
+tokenizers==0.21.1
+    # via transformers
+setuptools==69.0.2
+    # via torch
+torch==2.6.0
+    # via -r requirements.txt
+tqdm==4.67.1
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   transformers
+transformers==4.50.1
+    # via -r requirements.txt
+typing-extensions==4.12.2
+    # via
+    #   huggingface-hub
+    #   multidict
+    #   torch
+tzdata==2025.2
+    # via pandas
+urllib3==2.3.0
+    # via requests
+xxhash==3.5.0
+    # via datasets
+yarl==1.18.3
+    # via aiohttp
+accelerate==1.5.2
+psutil==5.9.8
+    # for accelerate
+anyio==4.9.0
+    # for transformers
+distro==1.9.0
+    # for transformers
+httpx==0.28.1
+    # for transformers
+jiter==0.9.0
+    # for transformers
+pydantic-core==2.33.0
+    # for transformers
+pydantic==2.11.0
+    # for transformers
+sniffio==1.3.1
+    # for httpx
+httpcore==1.0.7
+    # for httpx
+h11==0.14.0
+    # for httpx
+annotated-types==0.7.0
+    # for transformers
+typing-inspection==0.4.0
+    # for transformers
+exceptiongroup==1.2.2
+    # for transformers
+python-dotenv==1.1.0
+    # for loading env variables
+scipy==1.15.2
+    # for torch to get embeddings

finetuning_on_cpu/python_provider.sh

@@ -0,0 +1,30 @@
+#!/usr/bin/env bash
+# Generic Python test wrapper that handles cross-architecture execution
+
+# Get the directory where this script is located
+DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+
+# First argument is the Python module or script to run
+PYTHON_SCRIPT="$1"
+# Remaining arguments are passed to the Python script
+shift
+
+echo "Running Python test: $PYTHON_SCRIPT"
+echo "Python version:"
+python3 --version
+
+# Check if the argument is a file or a module
+if [[ -f "$PYTHON_SCRIPT" ]]; then
+    # Run as a script if it's a file
+    echo "Running as script"
+    python3 "$PYTHON_SCRIPT" "$@"
+else
+    # Run as a module if it's not a file
+    echo "Running as module"
+    python3 -m "$PYTHON_SCRIPT" "$@"
+fi
+
+# Capture and return the exit code
+exit_code=$?
+echo "Python test completed with exit code: $exit_code"
+exit $exit_code

finetuning_on_cpu/src/training/BUILD

@@ -0,0 +1,31 @@
+load("@rules_python//python:py_binary.bzl", "py_binary")
+load("@rules_shell//shell:sh_test.bzl", "sh_test")
+
+exports_files(
+    ["train_model.py"],
+    visibility=["//visibility:public"],  # Makes it visible to all packages
+)
+
+py_binary(
+    name="train_model",
+    srcs=["train_model.py"],
+    deps=[
+        "@pypi//torch",
+        "@pypi//transformers",
+        "@pypi//datasets",
+        "@pypi//accelerate",
+        "@pypi//psutil",
+    ],
+    visibility=["//visibility:public"],
+)
+
+sh_test(
+    name="training_test",
+    srcs=["//:python_provider.sh"],  # Use the python provider script from root
+    args=["src.training.train_model"],  # Pass the Python module path as an argument
+    data=[  # The Python script as a data dependency
+        "//src/training:train_model.py",
+    ],
+    # Optional: Set appropriate test size/timeout
+    size="large",
+)