<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bart-large-cnn-samsum-ElectrifAi_v8.1
This model is a fine-tuned version of philschmid/bart-large-cnn-samsum on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3222
- Rouge1: 55.3039
- Rouge2: 31.3218
- Rougel: 42.3951
- Rougelsum: 53.2394
- Gen Len: 108.9
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 27 | 1.3061 | 53.8018 | 30.0487 | 39.9195 | 52.1464 | 101.4333 |
| No log | 2.0 | 54 | 1.2995 | 54.2973 | 30.6364 | 42.0125 | 51.995 | 99.6 |
| No log | 3.0 | 81 | 1.3222 | 55.3039 | 31.3218 | 42.3951 | 53.2394 | 108.9 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.6.1
- Tokenizers 0.13.2
