1. Call the Scale or Shift functions when the parameter shift is 0.0F or scale is 1.0F.
2. Add the unit test for ScaleAndShift for the cases which parameter shift is 0.0F or scale is 1.0F.

Use the _CheckData without tolerance in some unit test to speedup the test process.

1. Use the _CheckData with tolerance in unit test to avoid the bug in "0.0F != -0.0F" case.
2, Clean the codes.

Optimize the MultiplyMe and DivMe functions when operate the scalar tensors.

1. Fix the bug in DivMe function which cannot handle the scalar tensor and broadcast case.
2. Clean the codes.
3. Fix minor errors.

1. Fix the bug in SubMe function which cannot handle the scalar tensor and broadcast case.
2. Clean the codes.

Fix the bug in SumMe function which cannot handle the scalar tensor and broadcast case.

1. Fix the bug in MultiplyMe function which cannot handle the scalar tensor and broadcast case.
2. Add broadcast multiply unit test.

Fix the bugs in FlushToMem functions.

Fix the bugs in FlushToMem functions.

1. Fix the bugs in SetDevice functions and FlushToMem functions.
2. Fix the minor errors.

1. CPU Support for Gather function.
2. Fix the bugs in GPU Gather.
2. Minor error fixed.

Update the link in manual.

Update the manual.

Update the Transformer codes.

1. Clean the codes.
2. Fix minor errors.

Clean the codes.

1. Fix minor errors.
2. Add some annotations.

Clean the codes and fix some warning issues.

Clean the codes.

Clean the codes.

1. Supporting efficient propagate and gradient accumulation for backward functions.
2. Update the setData functions.
3. Clean the codes.

1. Add boundary check for Gather function.
2. Minor error fixed.

Fix the bugs in CUDA 10.

Minor error fixed.

Fix the dimension bugs in Stack functions.

Minor error fixed.

Minor error fixed.

1. Fix the bugs for MatrixMul functions which may cause nan in some cases. (IMPORTANT)
2. Fix the minor errors.

1. Offer global setting for enableGrad.
2. Update XList.
3. Minor error fixed.

Minor error fixed.

Fix the GPU bug in Clip function.

1. Support X_INT dataType for Clip and ScaleAndShift function.
2. Minor error fixed.

1. Support X_INT dataType for Sum function.
2. Minor error fixed.