| import gradio as gr | |
| from transformers import AutoModelForSeq2SeqLM | |
| from transformers import AutoTokenizer | |
| model = AutoModelForSeq2SeqLM.from_pretrained('hackathon-pln-es/t5-small-spanish-nahuatl') | |
| tokenizer = AutoTokenizer.from_pretrained('hackathon-pln-es/t5-small-spanish-nahuatl') | |
| def predict(input): | |
| input_ids = tokenizer('translate Spanish to Nahuatl: ' + input, return_tensors='pt').input_ids | |
| outputs = model.generate(input_ids) | |
| outputs = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0] | |
| return outputs | |
| gr.Interface( | |
| fn=predict, | |
| inputs=gr.inputs.Textbox(lines=1, label="Input Text in Spanish"), | |
| outputs=[ | |
| gr.outputs.Textbox(label="Translated text in Nahuatl"), | |
| ], | |
| theme="peach", | |
| title='🌽 Spanish to Nahuatl Automatic Translation', | |
| description='This model is a T5 Transformer (t5-small) fine-tuned on 29,007 spanish and nahuatl sentences using 12,890 samples collected from the web and 16,117 samples from the Axolotl dataset. The dataset is normalized using "sep" normalization from py-elotl. For more details visit https://huggingface.co/hackathon-pln-es/t5-small-spanish-nahuatl', | |
| examples=[ | |
| 'hola', | |
| 'conejo', | |
| 'estrella', | |
| 'te quiero mucho', | |
| 'te amo', | |
| 'quiero comer', | |
| 'esto se llama agua', | |
| 'mi abuelo se llama Juan', | |
| 'te amo con todo mi corazón'], | |
| allow_flagging="manual", | |
| flagging_options=["right translation", "wrong translation", "error", "other"], | |
| flagging_dir="logs" | |
| ).launch(enable_queue=True) | |