flan-t5-large-medication-lists

This model is a fine-tuned version of google/flan-t5-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1217

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: 0.0003
• 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
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
5.8837	0.14	10	3.0384
2.3005	0.29	20	1.3933
6.7398	0.43	30	1.1939
1.1056	0.58	40	0.9215
0.5642	0.72	50	0.4442
0.6142	0.87	60	0.3734
0.2249	1.01	70	0.4756
11.7466	1.16	80	0.3282
0.1689	1.3	90	0.3007
0.2137	1.45	100	0.2909
0.1703	1.59	110	0.4483
16.8974	1.74	120	0.3103
0.1207	1.88	130	0.2340
0.2416	2.03	140	0.2377
0.1477	2.17	150	0.1635
0.1021	2.32	160	0.2011
8.3316	2.46	170	0.3211
14.5038	2.61	180	0.3357
0.0668	2.75	190	0.2597
12.083	2.9	200	0.2205
0.1218	3.04	210	0.1276
13.2802	3.19	220	0.1452
0.0587	3.33	230	0.1759
0.0714	3.48	240	0.1789
0.0819	3.62	250	0.1466
0.0461	3.77	260	0.1457
0.0597	3.91	270	0.1560
8.6414	4.06	280	0.1322
17.3547	4.2	290	0.1115
16.9251	4.35	300	0.1273
0.0516	4.49	310	0.1229
0.0335	4.64	320	0.1210
17.3581	4.78	330	0.1211
0.0234	4.93	340	0.1217

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1
• Datasets 2.14.5
• Tokenizers 0.13.3