SChem5Labels-google-t5-v1_1-large-inter_model-shuffle-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: 2.3184

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
20.749	1.0	25	23.9108
19.7523	2.0	50	22.3687
18.7945	3.0	75	18.8753
16.0524	4.0	100	12.6226
14.3498	5.0	125	9.9242
11.1053	6.0	150	9.2417
10.0075	7.0	175	9.1269
8.7375	8.0	200	8.9141
8.4414	9.0	225	8.8544
8.2513	10.0	250	8.8037
8.2066	11.0	275	8.7007
8.1662	12.0	300	8.5217
7.8864	13.0	325	8.2216
7.7552	14.0	350	7.9527
7.5072	15.0	375	7.7595
7.4867	16.0	400	7.6285
7.3527	17.0	425	7.5356
7.1715	18.0	450	7.4794
7.1983	19.0	475	7.4331
7.1102	20.0	500	7.3796
6.9301	21.0	525	5.8444
1.0577	22.0	550	0.9835
1.0083	23.0	575	0.9668
1.0127	24.0	600	0.9668
0.9964	25.0	625	0.9711
0.9749	26.0	650	0.9647
0.9807	27.0	675	0.9651
0.9811	28.0	700	0.9645
0.9935	29.0	725	0.9646
0.9805	30.0	750	0.9618
1.0088	31.0	775	0.9643
0.9957	32.0	800	0.9629
0.9886	33.0	825	0.9631
0.986	34.0	850	0.9624
0.9826	35.0	875	0.9633

Framework versions

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