gpt-2
gpt-2 copied to clipboard
nshepperd
Code for the paper "Language Models are Unsupervised Multitask Learners"
See https://github.com/nshepperd/gpt-2/issues/33
I could not train on smaller data sets, so I added an argument sample_batch_size to the train.py file, to make it possible to specify the sample size. Additionally, I changed...
I included a CLI argument to `encode.py` allowing to specify the model directory. The default was kept as `models`.
Теперь можно дообучать нейросеть без использования PYTHONPATH. Дополнил зависимости проекта, были указаны не все.
This file automates the model folder and file copying required after user fine-tunes the model using their own custom dataset. Usage: `python create_model.py -create mymodel -model 124M -run run2` This...
Using tensorflow 1.12.0 with the rest of the instructions leads to error "no attribute sort". A fix is to upgrade to tensorflow 1.15.0 (https://github.com/openai/gpt-2/issues/242)
How to train the GPT2-xl on TPU? And which TPU can be used to train? And what would be RAM size?
I'm not a coder or programmer, just a highschooler who wants to use GPT-2 to write a mythological fantasy light novel. I followed this tutorial from Alizarin 9 to download...
