huggingtweets/itemlabel

The model is based on a pre-trained GPT-2 which is fine-tuned on @itemlabel's tweets . Hyperparameters and metrics are recorded in the W&B training run for full transparency and reproducibility . The model suffers from the same limitations and bias as the model . You can use this model directly with a pipeline for text generation: pipeline('text-generation', model='huggingtweets/itemlabel') The model uses the following pipeline. The data is tracked with WandB artifacts at every step of the pipeline . The final model is logged and versioned at the end of training, and the model is versioned. It is based,