mkingopng / feedback_prize_pytorch Goto Github PK

Need to add TQDM progress bars

huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... To disable this warning, you can either: Avoid using tokenizers before the fork if possible, Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)

need to add Kfolds to HF approach

get this code to work on the large model

to deal with spelling mistakes, punctuation & grammar

transformers warning: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW() instead, or set no_deprecation_warning=True to disable this warning.

once I decide if I'm minimizing loss or maximizing f1, save epochs that set a new best for that metric

use wandb sweeps to optimize hyperparameters