there is a bug about 'get_ dropout_ mask_ for_cell' in recurrent.py
Open
- Dominant language
- Jupyter Notebook
- Stars
- 712
- Forks
- 77
- Avg merge
- 9h 39m
- Merged PRs (30d)
- 1
Description
The batch size of my program changes dynamically, but the function 'get_ dropout_ mask_ for_cell' in recurrent.py can only generate a fixed size of mask.
After run the code like ' inputs_i = inputs * dp_mask[0] ' an error occured.
Because my inputs shape like [5274,512] but dp_mask[0] like [9,512]. After the code initializes the dimension(9), the program will change the batch dimension to 5274.
So how can i fix it?
Contributor guide
Assessment
This issue has not been assessed yet.