bart-cnn-science-v3-e2-v4-e4-manual

This model is a fine-tuned version of theojolliffe/bart-cnn-science-v3-e2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1223
• Rouge1: 50.8519
• Rouge2: 30.3314
• Rougel: 31.5149
• Rougelsum: 48.4389
• Gen Len: 142.0

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: 2
• eval_batch_size: 2
• 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	42	0.9420	53.5234	33.6131	35.8383	51.1499	142.0
No log	2.0	84	0.9439	52.388	32.1451	35.2339	49.6554	142.0
No log	3.0	126	1.0321	56.2765	37.671	39.2693	53.5596	142.0
No log	4.0	168	1.1223	50.8519	30.3314	31.5149	48.4389	142.0

Framework versions

• Transformers 4.20.0
• Pytorch 1.11.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1