pegasus-large-finetuned-rahulver-summarization-pegasus-model

This model is a fine-tuned version of google/pegasus-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0906
• Rouge1: 61.2393
• Rouge2: 43.8277
• Rougel: 50.0054
• Rougelsum: 57.4674
• Gen Len: 114.6

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: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.3648	1.0	140	0.7201	50.0081	32.6454	39.3021	45.1602	125.7333
0.8502	2.0	280	0.6067	57.8678	41.5251	46.0694	54.1055	128.3333
0.5053	3.0	420	0.6642	58.3644	41.8619	47.6199	54.1639	108.9667
0.3469	4.0	560	0.7318	61.8988	45.7303	51.1928	57.9306	123.1667
0.2779	5.0	700	0.7274	62.9354	46.5	51.6431	59.2443	99.6333
0.2124	6.0	840	0.8618	63.8552	48.3846	53.3804	60.2718	111.2333
0.1864	7.0	980	1.0058	59.5675	42.4324	48.462	55.3498	108.4667
0.1691	8.0	1120	0.9984	60.1063	43.6022	49.7163	56.9865	130.2
0.1603	9.0	1260	1.0062	61.398	44.4507	50.2044	57.4447	99.0333
0.1674	10.0	1400	1.0906	61.2393	43.8277	50.0054	57.4674	114.6

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.7.1
• Tokenizers 0.13.2