Issues
- 2
- 3
Find small dimension typo in reading_comprehension/bidirectional_attention.py
#412 opened by felizxia - 1
Getting a nan softmax loss
#411 opened by bhavikajalli - 0
GatedAttention doesn't load correctly
#218 opened by matt-gardner - 1
evaluate_model() gives parsing error
#410 opened by mohitbadwal - 6
Use virtualenv instead of conda
#381 opened by schmmd - 3
Cannot run `bidaf_squad.json`
#408 opened by schmmd - 0
Better API around continuing training
#403 opened by matt-gardner - 4
- 1
Unpin Sphinx version
#363 opened by matt-gardner - 2
scripts for data processing?
#382 opened by dykang - 0
- 8
InputShape as a list is broken in latest keras (2.0.4)
#330 opened by avr248 - 2
- 1
Figure out if our `switch` function is broken
#354 opened by matt-gardner - 0
Allow loading already-indexed data
#352 opened by matt-gardner - 4
- 3
Provide an easy model ensembling API
#306 opened by matt-gardner - 1
Allow for non-Keras optimizers
#315 opened by matt-gardner - 9
- 0
Remove pin on pandas version
#346 opened by matt-gardner - 12
pip install is failing
#336 opened by znat - 3
Fix paths in the code to be OS independent
#335 opened by matt-gardner - 1
Adaptive batch sizes
#322 opened by matt-gardner - 0
- 0
Get random seed into the parameter file
#333 opened by matt-gardner - 1
- 3
Investigate using cache better in CircleCI
#329 opened by matt-gardner - 1
Enable batch-parallelization on multiple GPUs
#305 opened by matt-gardner - 3
Upgrade to python 3.6
#310 opened by matt-gardner - 0
- 9
Remove theano support
#311 opened by matt-gardner - 0
- 0
Better parameter logging
#300 opened by matt-gardner - 0
- 3
Make docs incorporate READMEs
#314 opened by matt-gardner - 0
- 0
- 0
Enhance DataIndexer
#271 opened by nelson-liu - 0
Different Character Encodings
#272 opened by nelson-liu - 0
Allow tokenizers to insert begin and end tokens
#276 opened by matt-gardner - 0
- 0
Easy profiling functionality
#309 opened by matt-gardner - 0
Figure out a good way to handle increasing the vocabulary of a pre-trained model
#312 opened by matt-gardner - 1
Make layers importable from the module
#274 opened by matt-gardner - 2
- 1
- 0
Collapsible Code Blocks for Training Logs
#269 opened by nelson-liu - 0
Tests are Flaky
#270 opened by nelson-liu - 0