redrussianarmy/gpt2-turkish-cased

The Turkish GPT-2 model was trained on various texts for Turkish . The model is meant to be an entry point for fine-tuning on other texts . It was possible to create byte-level BPE with Tokenizers library of Huggingface . Both PyTorch and Tensorflow compatible weights are available in the GPT2-Turkish model. The model itself can be used in this way: Using the model itself is possible to use the great Transformers Pipelines for generating text: pipeline('text-generation', model="redrussianarmy/gpt2-turkish-cased") pipeline('Text-generation') Pipeline('text' is pipe('text),