Support Matrix

FastEstimator	Python	TensorFlow	PyTorch	CUDA	Installation Instruction
Nightly	3.8-3.10	2.11.1	2.0.1	11.8	master branch
1.6 (recent stable)	3.8-3.10	2.11.1	2.0.1	11.8	r1.6 branch
1.5	3.7-3.9	2.9.1	1.10.2	11.0	r1.5 branch
1.4	3.6-3.8	2.4.1	1.7.1	11.0	r1.4 branch
1.3	3.6-3.8	2.4.1	1.7.1	11.0	r1.3 branch
1.2	3.6-3.8	2.4.1	1.7.1	11.0	r1.2 branch
1.1	3.6-3.8	2.3.0	1.6.0	10.1	r1.1 branch

1. Install Dependencies

• Install TensorFlow

  • Linux:

    pip install tensorflow==2.11.1
    

  • Mac (M1/M2): Please follow this installation guide

  • Windows: Please follow this installation guide

• Install PyTorch

  • CPU:

    bash pip install torch==2.0.1+cpu torchvision==0.15.2+cpu torchaudio==2.0.2+cpu -f https://download.pytorch.org/whl/cpu/torch_stable.html

• GPU:

  pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 torchaudio==2.0.2+cu118 -f https://download.pytorch.org/whl/cu118/torch_stable.html
  

• Extra Dependencies:

• Windows:

  • Install Build Tools for Visual Studio 2019 here.

  • Install latest Visual C++ redistributable here and choose x86 for 32 bit OS, x64 for 64 bit OS.

• Linux:

  apt-get install libglib2.0-0 libsm6 libxrender1 libxext6
  

• Mac:

  • Please follow this installation guide

2. Install FastEstimator

• Stable:

  pip install fastestimator
  

• Nightly (Linux/Mac):

  pip install fastestimator-nightly
  

Docker Hub

Docker containers create isolated virtual environments that share resources with a host machine. Docker provides an easy way to set up a FastEstimator environment. You can simply pull our image from Docker Hub and get started:

• Stable:

• GPU:

  docker pull fastestimator/fastestimator:latest-gpu
  

• CPU:

  docker pull fastestimator/fastestimator:latest-cpu
  

• Nighly:

• GPU:

  docker pull fastestimator/fastestimator:nightly-gpu
  

• CPU:

  docker pull fastestimator/fastestimator:nightly-cpu