Instructions to use Akheela/Model_generative with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ESPnet
How to use Akheela/Model_generative with ESPnet:
unknown model type (must be text-to-speech or automatic-speech-recognition)
- Notebooks
- Google Colab
- Kaggle
| from transformers import GPT2LMHeadModel, GPT2Tokenizer | |
| # Chargement du modèle GPT-2 et le tokenizer | |
| model_name = "gpt2" | |
| tokenizer = GPT2Tokenizer.from_pretrained(model_name) | |
| model = GPT2LMHeadModel.from_pretrained(model_name) | |
| #Le prompt | |
| prompt = "Au début du 21ème siècle, les humains ont découvert une nouvelle technologie" | |
| # Encodage du prompt pour l'entrée dans le modèle | |
| inputs = tokenizer.encode(prompt, return_tensors='pt') | |
| # Générer du texte par exemple 100 tokens | |
| output = model.generate( | |
| inputs, | |
| max_length=100, | |
| num_return_sequences=1, | |
| do_sample = True, | |
| temperature = 0.7, | |
| top_k = 50, | |
| top_p = 0.95 | |
| ) | |
| # Décodage de la génération en texte compréhensible | |
| generated_text = tokenizer.decode(output[0], skip_special_tokens=True) | |
| #Affichage | |
| print(generated_text) | |