theanets
theanets copied to clipboard
lmjohns3
Neural network toolkit for Python
# Issue Type [x] Bug (Typo) # Steps to Replicate and Expected Behaviour - Examine README.rst, docs/index.rst and observe `parantheses`, however expect to see `parentheses`. - Examine theanets/feedforward.py, theanets/graph.py and...
predicts 0.5, 0.5, 0.5, 0.5 should be 0,1,1,0 unless I'm missing the point?
First of all, thanks a lot for this clean piece of software :+1: I am using .itertrain() to train a deep autoencoder with a train and a validation dataset (numpy.ndarray)...
Hi, I think there's something wrong with the quick start example. I see rising accuracy but no real words: > sed for light, but only as an oi|ma;loaob1lu oh eoobol...
Attempting to use `theanets.recurrent.Text` on a UTF8 encoded corpus used to give an error ``` --------------------------------------------------------------------------- UnicodeEncodeError Traceback (most recent call last) /home/fl350/bachbot/scripts/theanet/theanet.py in () 24 with codecs.open(path, 'r', 'utf-8')...
If I understand correctly, recurrent network training in theanets currently uses `self.h_0` as the initial state (which defaults to `None`, implying all zeros) for every training example. This goes well...
We should include the value of the loss in the output from the feed_forward method.
Some operations are going to require 4D inputs and outputs (e.g., 3D convolutions). We should add models to support this, probably just a copy of the theanets/recurrent.py module ported to...
It would be very nice to have the [ROC AUC](http://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_auc_score.html) as classifier loss. Theanets has hidden the tensor logic from me so successfully, that I'm not able to write the...
several days ago, I used the theanets,Autoencoder to extract audio features,while the training processure stoped at the 1536th iteration,and one day passed, the iteration still showed at 1536th iteration.Can you...
