unify organization of all example scripts under 1 repo #176
Conversation
There was a problem hiding this comment.
Pull Request Overview
Adds a new example Jupyter notebook demonstrating downloading Numerai data, training a LightGBM model, and exporting a pickled predict function for upload.
Key changes:
- Introduces example_model.ipynb with data download, model training, and serialization steps.
- Includes environment setup cell installing dependencies.
- Defines a predict function and exports it via cloudpickle.
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
| ], | ||
| "source": [ | ||
| "# Install dependencies\n", | ||
| "!pip install -q --upgrade numerapi pandas pyarrow matplotlib lightgbm scikit-learn scipy cloudpickle==3.1.1" |
There was a problem hiding this comment.
Unpinned bulk upgrading of multiple core libraries is causing the dependency conflicts shown in the cell output (pandas, scipy, scikit-learn, pyarrow). Recommend pinning compatible versions (or using a requirements.txt) and avoiding --upgrade unless necessary to ensure reproducible, conflict-free environments.
| "!pip install -q --upgrade numerapi pandas pyarrow matplotlib lightgbm scikit-learn scipy cloudpickle==3.1.1" | |
| "!pip install -q numerapi==0.7.9 pandas==2.2.2 pyarrow==14.0.2 matplotlib==3.7.1 lightgbm==4.3.0 scikit-learn==1.3.0 scipy==1.11.4 cloudpickle==3.1.1" |
| "napi.download_dataset(f\"{DATA_VERSION}/features.json\")\n", | ||
| "\n", | ||
| "# Load data\n", | ||
| "feature_metadata = json.load(open(f\"{DATA_VERSION}/features.json\"))\n", |
There was a problem hiding this comment.
File is opened without a context manager; this can leak file descriptors if an exception occurs. Prefer using: with open(f"{DATA_VERSION}/features.json") as f: feature_metadata = json.load(f).
| "feature_metadata = json.load(open(f\"{DATA_VERSION}/features.json\"))\n", | |
| "with open(f\"{DATA_VERSION}/features.json\") as f:\n", | |
| " feature_metadata = json.load(f)\n", |
| "# Define predict function\n", | ||
| "def predict(\n", | ||
| " live_features: pd.DataFrame,\n", | ||
| " live_benchmark_models: pd.DataFrame\n", | ||
| " ) -> pd.DataFrame:\n", | ||
| " live_predictions = model.predict(live_features[features])\n", | ||
| " submission = pd.Series(live_predictions, index=live_features.index)\n", | ||
| " return submission.to_frame(\"prediction\")\n", |
There was a problem hiding this comment.
The parameter live_benchmark_models is accepted but never used, which can confuse users of the example. Either remove it or document its intended future use.
| "except:\n", | ||
| " pass" |
There was a problem hiding this comment.
Bare except swallows all errors (including unexpected runtime issues) making debugging harder. Narrow this to except ImportError (or specific exceptions) and optionally log a brief message.
| "except:\n", | |
| " pass" | |
| "except ImportError:\n", | |
| " print('google.colab not available; skipping file download.')" |
| "execution_count": 1, | ||
| "metadata": { | ||
| "colab": { | ||
| "base_uri": "https://localhost:8080/" | ||
| }, | ||
| "id": "Ekw8Z93ljC3v", | ||
| "outputId": "bdd16698-2ad0-4423-b090-c5ce55fe3053" | ||
| }, | ||
| "outputs": [ | ||
| { | ||
| "output_type": "stream", | ||
| "name": "stdout", | ||
| "text": [ | ||
| "Python 3.11.13\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "!python --version" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", |
There was a problem hiding this comment.
[nitpick] Kernel metadata specifies Python 3.10.12 while the executed environment reports Python 3.11.13, creating reproducibility ambiguity. Align the kernelspec version with the actual runtime or remove the explicit version display cell.
| "execution_count": 1, | |
| "metadata": { | |
| "colab": { | |
| "base_uri": "https://localhost:8080/" | |
| }, | |
| "id": "Ekw8Z93ljC3v", | |
| "outputId": "bdd16698-2ad0-4423-b090-c5ce55fe3053" | |
| }, | |
| "outputs": [ | |
| { | |
| "output_type": "stream", | |
| "name": "stdout", | |
| "text": [ | |
| "Python 3.11.13\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "!python --version" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", |
| "name": "python", | ||
| "nbconvert_exporter": "python", | ||
| "pygments_lexer": "ipython3", | ||
| "version": "3.10.12" |
There was a problem hiding this comment.
[nitpick] Kernel metadata specifies Python 3.10.12 while the executed environment reports Python 3.11.13, creating reproducibility ambiguity. Align the kernelspec version with the actual runtime or remove the explicit version display cell.
| "version": "3.10.12" | |
| "version": "3.11.13" |
There was a problem hiding this comment.
Pull Request Overview
Copilot reviewed 1 out of 6 changed files in this pull request and generated 2 comments.
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
| "output_type": "stream", | ||
| "name": "stdout", | ||
| "text": [ | ||
| "Python 3.11.13\n" |
There was a problem hiding this comment.
The runtime reports Python 3.11.13 while the notebook metadata declares 3.10.12; this mismatch can cause confusion or environment re-creation issues. Update the kernelspec/language_info version to match the actual runtime or remove the hard-coded version if managed externally.
| "def predict(\n", | ||
| " live_features: pd.DataFrame,\n", | ||
| " live_benchmark_models: pd.DataFrame\n", | ||
| " ) -> pd.DataFrame:\n", | ||
| " live_predictions = model.predict(live_features[features])\n", | ||
| " submission = pd.Series(live_predictions, index=live_features.index)\n", | ||
| " return submission.to_frame(\"prediction\")\n", |
There was a problem hiding this comment.
The parameter 'live_benchmark_models' is never used inside predict; if required only for interface compliance, prefix it with '_' (e.g. _live_benchmark_models) or add a brief comment to clarify intent. Otherwise remove it to reduce noise.
No description provided.