text_shortening_model_v38

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: 32.2806
• Rouge1: 0.0
• Rouge2: 0.0
• Rougel: 0.0
• Rougelsum: 0.0
• Bert precision: 0.6712
• Bert recall: 0.6737
• 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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

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
2.9479	1.0	145	6.0655	0.1154	0.0035	0.0998	0.0997	0.6949	0.7234	7.8649	46	2	47.0901	8.4084
3.2977	2.0	290	7.9855	0.0026	0.0	0.0026	0.0026	0.6628	0.6805	3.0	3	3	62.0	0.0
2.7673	3.0	435	18.0330	0.0	0.0	0.0	0.0	0.6716	0.677	1.0	1	1	62.0	0.0
2.7007	4.0	580	16.7534	0.0	0.0	0.0	0.0	0.6617	0.6651	1.0	1	1	62.0	0.0
2.6519	5.0	725	19.3665	0.0	0.0	0.0	0.0	0.6636	0.6599	1.0	1	1	62.0	0.0
2.6334	6.0	870	19.0112	0.0	0.0	0.0	0.0	0.6583	0.6639	1.0	1	1	62.0	0.0
2.5888	7.0	1015	20.8393	0.0	0.0	0.0	0.0	0.6602	0.6737	1.0	1	1	62.0	0.0
2.5665	8.0	1160	20.7588	0.0	0.0	0.0	0.0	0.6503	0.6688	1.0	1	1	62.0	0.0
2.546	9.0	1305	23.6869	0.0	0.0	0.0	0.0	0.6646	0.6703	1.0	1	1	62.0	0.0
2.5334	10.0	1450	26.1563	0.0	0.0	0.0	0.0	0.6693	0.6685	1.0	1	1	62.0	0.0
2.5194	11.0	1595	26.2698	0.0	0.0	0.0	0.0	0.6682	0.6743	1.0	1	1	62.0	0.0
2.5152	12.0	1740	30.3763	0.0	0.0	0.0	0.0	0.6582	0.6645	1.0	1	1	62.0	0.0
2.5005	13.0	1885	26.7690	0.0	0.0	0.0	0.0	0.6693	0.6597	1.0	1	1	62.0	0.0
2.4942	14.0	2030	26.8399	0.0	0.0	0.0	0.0	0.6655	0.6674	1.0	1	1	62.0	0.0
2.4766	15.0	2175	26.8788	0.0	0.0	0.0	0.0	0.6689	0.671	1.0	1	1	62.0	0.0
2.4712	16.0	2320	29.2279	0.0	0.0	0.0	0.0	0.6693	0.6669	1.0	1	1	62.0	0.0
2.46	17.0	2465	31.1020	0.0	0.0	0.0	0.0	0.6675	0.6655	1.0	1	1	62.0	0.0
2.4493	18.0	2610	31.4642	0.0	0.0	0.0	0.0	0.6655	0.6737	1.0	1	1	62.0	0.0
2.4419	19.0	2755	31.2733	0.0	0.0	0.0	0.0	0.6593	0.6629	1.0	1	1	62.0	0.0
2.4323	20.0	2900	32.2806	0.0	0.0	0.0	0.0	0.6712	0.6737	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