text_shortening_model_v45

This model is a fine-tuned version of facebook/bart-large-xsum on the None dataset. It achieves the following results on the evaluation set:

• Loss: 26.8982
• Rouge1: 0.0
• Rouge2: 0.0
• Rougel: 0.0
• Rougelsum: 0.0
• Bert precision: 0.6649
• Bert recall: 0.672
• Average word count: 1.0
• Max word count: 1
• Min word count: 1
• Average token count: 62.0
• % shortened texts with length > 12: 0.0

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: 0.0003
• 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bert precision	Bert recall	Average word count	Max word count	Min word count	Average token count	% shortened texts with length > 12
3.3791	1.0	83	6.7318	0.0982	0.0	0.0972	0.0969	0.6855	0.6599	1.2937	2	1	16.7619	0.0
2.8727	2.0	166	10.3841	0.0	0.0	0.0	0.0	0.674	0.6911	3.0	3	3	62.0	0.0
2.7805	3.0	249	10.0261	0.0345	0.0	0.0346	0.0345	0.6746	0.6819	2.0	2	2	62.0	0.0
2.7183	4.0	332	9.5191	0.0	0.0	0.0	0.0	0.673	0.6736	1.0	1	1	62.0	0.0
2.7086	5.0	415	10.4466	0.0	0.0	0.0	0.0	0.6568	0.6648	1.0	1	1	62.0	0.0
2.6474	6.0	498	13.9665	0.0	0.0	0.0	0.0	0.6641	0.6709	1.0	1	1	62.0	0.0
2.63	7.0	581	13.3621	0.0	0.0	0.0	0.0	0.6457	0.6701	1.0	1	1	62.0	0.0
2.5998	8.0	664	13.0602	0.0	0.0	0.0	0.0	0.6618	0.6672	1.0	1	1	62.0	0.0
2.5689	9.0	747	15.0760	0.0	0.0	0.0	0.0	0.6591	0.6651	1.0	1	1	62.0	0.0
2.5508	10.0	830	15.6936	0.0	0.0	0.0	0.0	0.6649	0.6716	1.0	1	1	62.0	0.0
2.5298	11.0	913	16.8446	0.0	0.0	0.0	0.0	0.6604	0.6648	1.0	1	1	62.0	0.0
2.5091	12.0	996	21.0673	0.0	0.0	0.0	0.0	0.6721	0.6702	1.0	1	1	62.0	0.0
2.5019	13.0	1079	25.5628	0.0	0.0	0.0	0.0	0.6605	0.67	1.0	1	1	62.0	0.0
2.4826	14.0	1162	25.1203	0.0	0.0	0.0	0.0	0.6725	0.6666	1.0	1	1	62.0	0.0
2.4693	15.0	1245	26.8982	0.0	0.0	0.0	0.0	0.6649	0.672	1.0	1	1	62.0	0.0

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3