fix: add the momentum parameter's implemention of SGD

fix: add the gradient of the tf.gradient opr

feat: add adjust_contrast, adjust_hue, combined_non_max_suppression, crop_and_resize image oprs

fix: revise np.amin, np.amax and add np.argmin

Fix: model.load_weights

feat: add Bidirectional layer

fix: make the initialization of the layer's name correct

fix: ignore the LSTMLoad test

fix: fix the bug of load LSTM model and add test

Thanks for your contribution.

fix flip_left_right run bug

fix same bug

the bool to tensor has a bug, if in init the training is False, the program not start.

fix: exp moved to tf.math.cs

fix: Fix the problem where the  model.save_weights() can't save multi-layer weight

Add a new optimizer