[Wait for #2591] [ TEST ] Add more unittest and fixes for mixed precision

[jijoongmoon]

In this PR

This PR includes more unittest and fixes for mixed precision. . Model Unittest . 2 fc layer which generates NaN or Inf Gradient from Troch. . MSE Loss and Check the whole procedure of the mixed precision training. . Even if the FC model only has one weight, it is good enough to validate the mixed precision. . Torch model also works similar way to NNTrainer. . Some fixes about the execution order of applying gradient when the mixed precision is on. . Update SGD to support Mixed Precision training

Changes proposed in this PR:

• Added TOC generator for README.md

Resolves:

Self evaluation:

1. Build test: [X]Passed [ ]Failed [ ]Skipped
2. Run test: [X]Passed [ ]Failed [ ]Skipped

Signed-off-by: jijoong.moon [email protected]