textgenrnn
textgenrnn copied to clipboard
Easily train your own text-generating neural network of any size and complexity on any text dataset with a few lines of code.
Hello. When I am trying to train a new model on google colab. I run into the following error: ``` Training new model w/ 3-layer, 128-cell LSTMs Training on 125,286...
utf-8' codec can't decode byte 0xa4 in position 14: invalid start byte Full code is attached in Word file with csv file too [mcqs.csv](https://github.com/minimaxir/textgenrnn/files/7950821/mcqs.csv) [quiz.docx](https://github.com/minimaxir/textgenrnn/files/7950823/quiz.docx)
```Training new model w/ 3-layer, 128-cell Bidirectional LSTMs Training on 237,699 character sequences. Epoch 1/20 --------------------------------------------------------------------------- UnknownError Traceback (most recent call last) in () 18 max_length=model_cfg['max_length'], 19 dim_embeddings=100, ---> 20...
Previously, textgenrnn could work only if 1) Tensorflow < 2.5.0 2) Keras == 2.2.0 (or Keras
There is no 'open' button above the image - I guess Google Drive changed something on their end. The link should be changed to point directly to an .ipynb hosted...
This is a very useful task and many works are implemented based on it! However, when I use this repo to compare the performance of my work, I find that...
When trying to tokenize a dataset, it fails with either the error `Error: new-line character seen in unquoted field - do you need to open the file in universal-newline mode?`...
Output is: ``` 1,016,957 texts collected. Training new model w/ 2-layer, 128-cell LSTMs Training on 39,616,040 character sequences. Epoch 1/1 Traceback (most recent call last): File "train.py", line 7, in...
This may be just my setp, but installation of this package fails telling me it can't find TensorFlow 2.1.1.
The lib is great! I've managed to train a model and save the associated data Using the code; ```python from textgenrnn import textgenrnn textgen = textgenrnn() textgen.train_from_file('names.txt', new_model=True, word_level=True, num_epochs=100)...