| { | |
| "model_type": "t5", | |
| "architectures": [ | |
| "T5ForConditionalGeneration" | |
| ], | |
| "task": "summarization", | |
| "training_params": { | |
| "epochs": 10, | |
| "batch_size": 8, | |
| "learning_rate": 2e-05, | |
| "warmup_steps": 1000, | |
| "focal_loss": { | |
| "alpha": 1, | |
| "gamma": 2 | |
| } | |
| } | |
| } |
| { | |
| "model_type": "t5", | |
| "architectures": [ | |
| "T5ForConditionalGeneration" | |
| ], | |
| "task": "summarization", | |
| "training_params": { | |
| "epochs": 10, | |
| "batch_size": 8, | |
| "learning_rate": 2e-05, | |
| "warmup_steps": 1000, | |
| "focal_loss": { | |
| "alpha": 1, | |
| "gamma": 2 | |
| } | |
| } | |
| } |