SChem5Labels-google-t5-v1_1-large-intra_model-dataset-frequency-human_annots_str

This model is a fine-tuned version of google/t5-v1_1-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: nan

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: 0.0001
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss
20.6291	1.0	25	25.1686
19.5807	2.0	50	23.6946
18.3885	3.0	75	16.4506
15.6621	4.0	100	10.3654
14.1536	5.0	125	9.6059
11.1395	6.0	150	9.4016
10.234	7.0	175	9.1651
8.6134	8.0	200	8.8517
8.4527	9.0	225	8.7267
8.1685	10.0	250	8.6661
8.1419	11.0	275	8.5700
8.1851	12.0	300	8.4129
7.9855	13.0	325	8.1392
7.5843	14.0	350	7.8676
7.4604	15.0	375	7.6409
7.2433	16.0	400	7.4793
7.1782	17.0	425	7.3690
6.9756	18.0	450	7.2904
6.8602	19.0	475	7.2253
6.9456	20.0	500	7.1670
6.5134	21.0	525	0.6126
0.8013	22.0	550	0.5889
0.6372	23.0	575	0.5655
0.6054	24.0	600	0.5584
0.6018	25.0	625	0.5609
0.5884	26.0	650	0.5572
0.5887	27.0	675	0.5585
0.5919	28.0	700	0.5575
0.5908	29.0	725	0.5570
0.5859	30.0	750	0.5510
0.6028	31.0	775	0.5558
0.594	32.0	800	0.5491
0.5834	33.0	825	0.5556
0.5849	34.0	850	0.5515
0.5791	35.0	875	0.5519
0.5863	36.0	900	0.5553
0.5897	37.0	925	0.5486
0.582	38.0	950	0.5555
0.579	39.0	975	0.5519
0.5816	40.0	1000	0.5530
0.5822	41.0	1025	0.5483
0.5838	42.0	1050	0.5548
0.5767	43.0	1075	0.5479
0.5757	44.0	1100	0.5498
0.5911	45.0	1125	0.5515
0.5665	46.0	1150	0.5556
0.5773	47.0	1175	0.5518
0.5727	48.0	1200	0.5499

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1