PyTorch - AUTOMATIC MIXED PRECISION
FizzBuzz, inspired from the excellent blog post Joel Grus - Fizz Buzz in Tensorflow PyTorch implementation borrowed heavily from fizz-buzz-pytorch
This is meant to be a simple example of Automatic Mixed Precision (AMP) training in PyTorch 1.6. I wrote it because many examples were too simple or too complex. Hopefully it can provide a little amuzement and show how you may get faster training time taking advantage of AMP.
- PyTorch 1.6
- Numpy
- Colorama