bart-cnn-science-v3-e1-v4-e6-manual

This model is a fine-tuned version of theojolliffe/bart-cnn-science-v3-e1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4513
• Rouge1: 51.4471
• Rouge2: 31.5595
• Rougel: 31.7717
• Rougelsum: 49.4999
• 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: 6
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	42	1.0691	51.1883	31.2479	33.7004	48.9571	142.0
No log	2.0	84	1.0883	51.7634	29.8573	30.7155	49.3378	142.0
No log	3.0	126	1.2355	52.9606	31.3539	33.5131	49.9275	142.0
No log	4.0	168	1.3430	52.2108	32.7896	34.65	50.4271	139.1
No log	5.0	210	1.3963	51.5335	30.4157	31.5759	49.6904	142.0
No log	6.0	252	1.4513	51.4471	31.5595	31.7717	49.4999	142.0

Framework versions

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