SChem5Labels-google-t5-v1_1-large-inter_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
20.7769	1.0	25	24.1676
19.3503	2.0	50	20.2926
17.4487	3.0	75	12.6689
14.6317	4.0	100	10.4484
12.7323	5.0	125	9.5200
10.6392	6.0	150	9.2853
9.777	7.0	175	9.0216
8.5266	8.0	200	8.8159
8.4237	9.0	225	8.7335
8.1749	10.0	250	8.6666
8.1341	11.0	275	8.5495
8.1705	12.0	300	8.4080
7.9849	13.0	325	8.1718
7.5745	14.0	350	7.8604
7.4689	15.0	375	7.6465
7.2368	16.0	400	7.4584
7.1852	17.0	425	7.3699
6.9957	18.0	450	7.3086
6.8921	19.0	475	7.2569
7.0036	20.0	500	7.2094
6.8411	21.0	525	7.1594
6.6867	22.0	550	7.0519
1.1614	23.0	575	0.6432
0.6204	24.0	600	0.5628
0.6131	25.0	625	0.5761
0.591	26.0	650	0.5582
0.5949	27.0	675	0.5605
0.6009	28.0	700	0.5616
0.5916	29.0	725	0.5660
0.5927	30.0	750	0.5571
0.6041	31.0	775	0.5556
0.5988	32.0	800	0.5530
0.5894	33.0	825	0.5822
0.5899	34.0	850	0.5548
0.5813	35.0	875	0.5542
0.5901	36.0	900	0.5518
0.595	37.0	925	0.5502
0.585	38.0	950	0.5515
0.5837	39.0	975	0.5525
0.5834	40.0	1000	0.5499
0.5858	41.0	1025	0.5490
0.5892	42.0	1050	0.5498
0.5814	43.0	1075	0.5478
0.5779	44.0	1100	0.5492
0.5955	45.0	1125	0.5496
0.5711	46.0	1150	0.5535
0.5795	47.0	1175	0.5511
0.5783	48.0	1200	0.5527

Framework versions

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