Jindřich questions

Jindřich answers

• Huggingface GPT2 loss understanding
• Running huggingface Bert tokenizer on GPU
• can I train(optimize) on f1 score loss with pytorch
• Using fasttext pre-trained models as an Embedding layer in Keras
• Predict over a whole dataset using Transformers
• How many characters can be input into the "prompt" for GPT-2
• Choose vocabulary size of tokenizer
• Must the vocab size must math the vocab_size in bert_config.json exactly?
• How to untokenize BERT tokens?
• Changing config and loading Hugging Face model fine-tuned on a downstream task