Add ruff as linter and formatter & codex

.pre-commit-config.yaml

@@ -0,0 +1,12 @@
+repos:
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  # Ruff version.
+  rev: v0.14.8
+  hooks:
+    # Run the linter.
+    - id: ruff-check
+      types_or: [ python, pyi ]
+      args: [ --fix ]
+    # Run the formatter.
+    - id: ruff-format
+      types_or: [ python, pyi ]

.vscode/settings.json

@@ -0,0 +1,10 @@
+{
+  "[python]": {
+    "editor.formatOnSave": true,
+    "editor.codeActionsOnSave": {
+      //"source.fixAll": "explicit",
+      "source.organizeImports": "explicit"
+    },
+    "editor.defaultFormatter": "charliermarsh.ruff"
+  }
+}

README.md

@@ -100,9 +100,10 @@ This project uses `uv` for managing the python virtual environment. To install u
 curl -LsSf https://astral.sh/uv/install.sh | sh
 ```
 
-To create and activate the virtual environment, run the following:
+To create and activate the virtual environment and setup git hooks (for auto linting and formatting), run the following:
 ```bash
 uv sync --locked
+uv run pre-commit install
 ```
 
 ## Run instructions
@@ -143,12 +144,40 @@ The last command will start the MLflow GUI at your local host loopback on port 5
 
 ![mlflow_tune](./docs/images/mlflow_tune.png)
 
-## Future Work
+# Contributing
+
+## Linting and Formatting
+We are using [ruff](https://docs.astral.sh/ruff/) as a the linter and formatter. Ruff can be used with:
+
+```bash
+# Get check results
+uv run ruff check .
 
-For model improvements I'm interested in switching to a flash attention layer which can accelerate training
+# Get check results and apply fixes (recommend review these with git diff)
+uv run ruff check . --fix
 
-I also want to build up per team models, do ANOVA, then highlight specific players that are giving tells on coverages
+# Run formatter
+uv run ruff format .
+```
 
-Want to try out the model on a non-binary problem and try distinguishing between distinct coverage types
+Note, if you followed instructions above to setup the git hook, ruff will run the linter (with auto fixing on) and formatting as a pre-commit git hook.
 
-See if I can identify blitzer (is the blitzer player identified in the dataset?)
+## Future Work
+- Data
+  - Incoporate the additional NGS data which likely will have the largest returns in increasing accuracy
+  - Look into additional data augmentation techniques
+- Model
+  - Consider model improvements in transformer arch
+  - Build per team model, do ANOVA, then highlight specific players that are giving tells on coverages
+  - Build multiclass output model instead of binary classification
+  - See if I can build a model to determine blitzer like red circle AWS at post snap
+  - Add in confusion matrix, ROC curve, and prediction post snap accuracy to automatic predictions and not just in notebook
+- Infra
+  - Profiler
+    - Add in PyTorch profiler with new tensorboard (new thing, this is deprecated) traces into Chrome UI, built into CI for profiler
+  - Optimization
+    - Switch to flash attention layer which can accelerate training
+  - General
+    - Get docker working and implemented in CI
+    - Add git hook for ruff fix
+    - Look back on mypy for type