python-textualheatmap
python-textualheatmap copied to clipboard
Error when multiplying prediction logits by output mask tensor
I'm using TF 2.4 and this line predict_mask_correct_token = tf.reduce_sum(predict * output_mask_tensor)
gives error TypeError: Cannot convert 'logits' to EagerTensor of dtype float
Any ideas?
Have the same problem, anyone can help us?