This model is a fine-tuned version of GPT-2 using 21,832 tweets from 12 twitter users with very strong opinions about the United States Men’s National Team. Additionally, BOTHALTEROUT can create some problematic results based upon the tweets used to generate the model. The following hyperparameters were used during training: — learningrate: 0.0001372 — trainbatchsize: 1 — evalbatchsize: 8 — seed: 42 — optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 — lrschedulertype: linear — numepochs: 1 — Transformers 4.19.2 — Pytorch 1.11.0+cu113 — Datasets 2.2.2 — Tokenizers 0.12.1 Built by Eliot McKinley based upon HuggingTweets by Boris Dayama
Модальности:
Генерация текста
Задача: Генерация текста
Автор: etmckinley
Теги: tensorboard, gpt2, text-generation-inference, endpoints_compatible
Лайков: 2 | Загрузок: 21
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.