daquang

I believe you still have some errors. In your newest version, you have these lines: model.add(BiDirectionLSTM(word_vec_len, 50, output_mode='concat'), return_sequences=True) model.add(BiDirectionLSTM(100, 24, output_mode='sum'), return_sequences=True) After changing these two lines as follows,...

Hi Patrick, Thank you for your interest in FactorNet. Unfortunately, I have left academia and I am no longer supporting this package. However, I have been developing new packages, GenomeLoader...

I actually wrote PillowNet while I was in my postdoc under the MIT license, so it is freely available. I expect the results are not too different from FactorNet, and...

Have you tested Bidirectional? I get errors when I try using it on the imdb example data. For example, it complains about return_sequences not being defined.

Still getting errors. What version of theano and keras are you using?

I'm not using the latest github versions, just the latest stable versions. Theano 0.7 and Keras 0.1.3. Here is my error: Traceback (most recent call last): File "imdb_brnn.py", line 67,...

Can you describe to me your dataset and show me the output motifs and your command line? Motif discovery is a bit more art than science. I suspect you may...

That does sound like a very difficult dataset to analyze. I'm not sure if any motif discovery algorithm can get high quality motifs from it. Have you tried centrimo?

@milesroberts-123 It looks like you're using your own negative sequences. Can you try using a dinucleotide shuffled version instead? A lot of repetitive motifs suggests you have a lot of...

@milesroberts-123 Hmm, why don't you try going into the erase_annoying_sequences script and modify the following line: annoying_subsequences = [ 'AAAAAA', 'TTTTTT', 'CCCGCCC', 'GGGCGGG' ] and add 'CCCCC' and 'GGGGG' to...