onnxruntime/requirements-training.txt

cerberus
flatbuffers
h5py
numpy >= 1.16.6
onnx
packaging
protobuf >= 3.20.0, <= 3.21.0
sympy
setuptools>=41.4.0
Add missing Python dependencies for ORT training (#7104) * Add missing Python dependencies for training cerberus - option parsing h5py - checkpoint onnx - model proto packaging/sympy - symbolic shape inference * Separate requirements.txt for inference and training Python packages. 2021-03-24 01:43:19 +00:00			`cerberus`
			`flatbuffers`
			`h5py`
			`numpy >= 1.16.6`
			`onnx`
			`packaging`
Change protobuf pin in training requirements (#13596) 2022-11-09 17:37:41 +00:00			`protobuf >= 3.20.0, <= 3.21.0`
Add post-install command to build PyTorch CPP extensions from within onnxruntime package (#8027) ORTModule requires two PyTorch CPP extensions that are currently JIT compiled. The runtime compilation can cause issues in some environments without all build requirements or in environments with multiple instances of ORTModule running in parallel This PR creates a custom command to compile such extensions that must be manually executed before ORTModule is executed for the first time. When users try to use ORTModule before the extensions are compiled, an error with instructions are raised PyTorch CPP Extensions for ORTModule can be compiled by running: python -m onnxruntime.training.ortmodule.torch_cpp_extensions.install Full build environment is needed for this 2021-06-29 01:11:58 +00:00			`sympy`
			`setuptools>=41.4.0`