Add basic LSTM tutorial #55
Conversation
|
|
||
| return x, x_mask, labels | ||
|
|
||
| def load_data(path="/data/lisatmp3/chokyun/tweets_sa/imdb/aclImdb/imdb.pkl", n_words=100000, valid_portion=0.1): |
There was a problem hiding this comment.
That path should be changed to point to the location of the downloaded data.
There was a problem hiding this comment.
Good point. That's a copy-paste artefact
|
I removed the commented out code. Concerning the other points, I am not certain why things are implemented this way. This is Cho and other' LSTM code that Fred asked me to put in a tutorial. I only removed external dependencies to sklearn and scipy and wrote a tutorial around it. I can pep8-ize it but further clean up/refactoring/documenting would probably be more efficiently carried out by the original authors. |
|
@carriepl, it'll be nice if you can pep8-ize it. Optimization/clean-up/refactoring may/should happen later by me or others. |
|
pep8 was fixed. |
|
I merge it now to annonce this example. We can polish it later. |
Add basic LSTM tutorial
Add basic LSTM tutorial Former-commit-id: 664b4cc
No description provided.