Installation ============ Requirements ------------ * Python 3.10 or 3.11 (TensorFlow 2.14.1 compatibility requirement) * TensorFlow 2.14.1 (automatically installed) .. note:: Python 3.12+ is not currently supported due to TensorFlow availability constraints. Standard Installation --------------------- We strongly recommend installing pranaam inside a Python virtual environment. (see `venv documentation `_) Install pranaam using pip: .. code-block:: bash pip install pranaam This installs TensorFlow 2.14.1, which is known to work correctly with the models. Installation Options -------------------- For development work: .. code-block:: bash pip install -e .[dev] For testing: .. code-block:: bash pip install -e .[test] For documentation building: .. code-block:: bash pip install -e .[docs] For all optional dependencies: .. code-block:: bash pip install -e .[all] TensorFlow Compatibility ------------------------ The package requires TensorFlow 2.14.1 with Keras 2.14.0 for model compatibility. If you encounter compatibility issues: .. code-block:: bash pip install 'pranaam[tensorflow-compat]' Model Downloads --------------- Models are automatically downloaded from Harvard Dataverse (306MB) and cached locally on first use. Ensure you have: * Stable internet connection * At least 500MB free disk space * Unrestricted access to dataverse.harvard.edu Verification ------------ Test your installation: .. code-block:: python import pranaam result = pranaam.pred_rel("Shah Rukh Khan") print(result) If successful, you should see a pandas DataFrame with prediction results. Troubleshooting --------------- Common Issues ~~~~~~~~~~~~~ **TensorFlow/Keras Compatibility Errors** Error: ``"Keras 3 only supports V3 .keras files and legacy H5 format files"`` Solution: Install with ``pip install 'pranaam[tensorflow-compat]'`` **Model Download Issues** Error: Network timeouts or download failures Solution: Check internet connection, models are large (306MB) **Import Errors** Error: ``pkg_resources`` deprecation warnings Solution: Already fixed in v0.1.0 (uses ``importlib.resources``) For additional help, please check our `GitHub Issues `_.