README.md

metadata

license: apache-2.0
base_model: google/flan-t5-small
tags:
  - generated_from_trainer
model-index:
  - name: medication-lists
    results: []

medication-lists

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

• Loss: 0.1133

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: 16
• eval_batch_size: 16
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.6316	0.46	50	0.6287
0.6403	0.93	100	0.3378
0.4213	1.39	150	0.2460
0.3452	1.85	200	0.2184
0.306	2.31	250	0.1903
0.2634	2.78	300	0.1807
0.2423	3.24	350	0.1630
0.2224	3.7	400	0.1599
0.2107	4.17	450	0.1522
0.1922	4.63	500	0.1515
0.1887	5.09	550	0.1394
0.1821	5.56	600	0.1414
0.1705	6.02	650	0.1378
0.1602	6.48	700	0.1330
0.1579	6.94	750	0.1300
0.1497	7.41	800	0.1282
0.1534	7.87	850	0.1277
0.147	8.33	900	0.1274
0.1395	8.8	950	0.1204
0.1361	9.26	1000	0.1235
0.1353	9.72	1050	0.1210
0.1303	10.19	1100	0.1220
0.132	10.65	1150	0.1232
0.1262	11.11	1200	0.1193
0.1228	11.57	1250	0.1229
0.1261	12.04	1300	0.1215
0.1204	12.5	1350	0.1163
0.12	12.96	1400	0.1189
0.11	13.43	1450	0.1173
0.1183	13.89	1500	0.1149
0.108	14.35	1550	0.1178
0.1122	14.81	1600	0.1150
0.1126	15.28	1650	0.1157
0.112	15.74	1700	0.1152
0.1046	16.2	1750	0.1156
0.1057	16.67	1800	0.1138
0.1067	17.13	1850	0.1129
0.1078	17.59	1900	0.1140
0.1043	18.06	1950	0.1135
0.1033	18.52	2000	0.1138
0.1017	18.98	2050	0.1140
0.102	19.44	2100	0.1125
0.1012	19.91	2150	0.1133

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu121
• Datasets 2.11.0
• Tokenizers 0.14.1