text-generator
text-generator copied to clipboard
ChunML
Having fun with LSTMs by creating a simple text generator.
Fixed a bug where the character list generated upon data loading would return different ordered results every time it's called resulting in non-reproducible behaviour. This leads to the generation of...
Hi Trung, I am trying to implement your code, but there is one line I couldn't understand. In the RNN_utils.py, Line 38 and Line 45. I think "X[i] = input_sequence"...
Tried to load the model and continue training, it seems like it starts with high loss and the model seems to be totally untrained (start over at epoch 0). The...
I'm trying to implement the recurrent neural network using Keras but with Tensorflow as the back-end instead of Theano. Is this possible? Any gotchas or tips/tricks are appreciated. Thank you!
Traceback (most recent call last): ``` ],,NNVEEE,EKKkkkEKkkkqqqqqqqssqsq---rrrr!!!zKKrrrPkkzzKKkkrrKkkkrkKkkqqqqqqqssqsq---rrrr!!!zKKrrrPkk File "C:/Users/username/AppData/Local/Programs/Python/Python36/text-generator-master/text-generator-master/recurrent_keras.py", line 54, in model.load_weights('') File "C:\Users\usernameAppData\Local\Programs\Python\Python36\lib\site-packages\keras\models.py", line 721, in load_weights f = h5py.File(filepath, mode='r') File "C:\Users\username\AppData\Local\Programs\Python\Python36\lib\site-packages\h5py\_hl\files.py", line 269, in __init__ fid...
All the lines having smth like data[i*seq_length:(i+1)*seq_length] actually get 1/seq_length of the training data, as they convert data to disjoint sequences of seq_length length. Proposing a small enhancement - to...
Running the code gives me the following error: > Data length: 3204677 characters > Vocabulary size: 94 characters > Traceback (most recent call last): > File "recurrent_keras.py", line 36, in...
Thanks for sharing. Are the default parameters in the code the same as those used in your [post](https://chunml.github.io/ChunML.github.io/project/Creating-Text-Generator-Using-Recurrent-Neural-Network/)? And how large is your training data (Harry Potter series)?
