Conclusion
06 Oct 2018This was an amazing experience, but today I pulled the plug.
- San Francisco
- GitHub
- Stackoverflow
- YouTube
-
Google Scholar
Recent Posts
Dropout Experiment with Only Popular Words
05 Nov 2017Continuing from the previous experiment eliminating uncommon words, adjusting the Dropout produces an effect, but the improvement is not significant enough.
Eliminating Uncommon Words Makes Things Worse
03 Nov 2017Eliminating sentences with uncommon words (i.e. those that only appear once) from the dataset does not improve the accuracy of the training, for reasons unknown.
Reducing the Validation and Test datasets
20 Oct 2017Depending on the situation, using smaller Validation and Test datasets can leave more data for training, and reduce time spent computing accuracy, without any impact on determining accuracy.
No hablo español
16 Oct 2017Remove non-English text from the data set.
He normal makes things worse
01 Oct 2017Disappointing that a fancy, modern initialization technique not only didn’t help, but made things worse
Goldilocks batch size
01 Oct 2017Some batch sizes are too big and some are too small. It’s important to find the one that is just right.