SChem5Labels-google-t5-v1_1-large-intra_model-frequency-model_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: 0.8369

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
19.8258	1.0	25	23.6474
19.0315	2.0	50	22.0370
18.08	3.0	75	19.1087
16.4078	4.0	100	11.6693
14.5372	5.0	125	10.0089
12.1759	6.0	150	9.5651
10.8249	7.0	175	9.2475
9.4751	8.0	200	8.9311
8.801	9.0	225	8.6771
8.1126	10.0	250	8.5237
7.9399	11.0	275	8.4068
7.9146	12.0	300	8.3324
7.9766	13.0	325	8.2182
7.6203	14.0	350	8.0454
7.5088	15.0	375	7.7369
7.2191	16.0	400	7.4618
7.0805	17.0	425	7.2855
6.8971	18.0	450	7.1672
6.8954	19.0	475	7.0791
6.7074	20.0	500	7.0220
6.6851	21.0	525	6.9700
6.6409	22.0	550	6.9230
6.5565	23.0	575	6.8672
6.4106	24.0	600	6.8143
5.2007	25.0	625	2.4787
0.9209	26.0	650	0.7582
0.8058	27.0	675	0.7280
0.7899	28.0	700	0.7278
0.7875	29.0	725	0.7233
0.7813	30.0	750	0.7216
0.7621	31.0	775	0.7209
0.7893	32.0	800	0.7156
0.7727	33.0	825	0.7116
0.7562	34.0	850	0.7143
0.7639	35.0	875	0.7136
0.7553	36.0	900	0.7087
0.7382	37.0	925	0.7072
0.7361	38.0	950	0.7106
0.7469	39.0	975	0.7059
0.7516	40.0	1000	0.7074
0.7478	41.0	1025	0.7051
0.7367	42.0	1050	0.7096
0.7417	43.0	1075	0.7057
0.7434	44.0	1100	0.7056
0.7433	45.0	1125	0.7022
0.7538	46.0	1150	0.7024
0.7246	47.0	1175	0.7004
0.7418	48.0	1200	0.7014
0.7469	49.0	1225	0.7038
0.7184	50.0	1250	0.6997
0.7459	51.0	1275	0.6998
0.716	52.0	1300	0.7008
0.7269	53.0	1325	0.7015
0.7354	54.0	1350	0.6979
0.7209	55.0	1375	0.6976
0.728	56.0	1400	0.6948
0.7201	57.0	1425	0.6986
0.7228	58.0	1450	0.6957
0.7254	59.0	1475	0.6996
0.7252	60.0	1500	0.6961
0.7116	61.0	1525	0.6987

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.6.1
• Tokenizers 0.14.1