metadata
license: mit
base_model: facebook/bart-large-cnn
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: video_transcript_summary
results: []
video_transcript_summary
This model is a fine-tuned version of facebook/bart-large-cnn on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.9619
- Rouge1: 0.3774
- Rouge2: 0.1256
- Rougel: 0.2607
- Rougelsum: 0.2615
- Gen Len: 69.9412
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 5 | 2.1870 | 0.3207 | 0.0851 | 0.2147 | 0.2132 | 75.6471 |
| No log | 2.0 | 10 | 2.0146 | 0.3715 | 0.1267 | 0.2567 | 0.2563 | 68.7059 |
| No log | 3.0 | 15 | 1.9649 | 0.3759 | 0.1432 | 0.2809 | 0.2818 | 66.5882 |
| No log | 4.0 | 20 | 1.9619 | 0.3774 | 0.1256 | 0.2607 | 0.2615 | 69.9412 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.1+cu118
- Datasets 2.20.0
- Tokenizers 0.19.1