README.md

metadata

library_name: transformers
tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: WhartonDS_ClsModelTest
    results: []

WhartonDS_ClsModelTest

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

• Loss: 0.0610
• Auc Roc: 0.9980
• F1: 0.9824

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: 5e-05
• train_batch_size: 128
• eval_batch_size: 64
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 70

Training results

Training Loss	Epoch	Step	Validation Loss	Auc Roc	F1
0.6934	1.0	24	0.6987	0.4096	0.4188
0.6806	2.0	48	0.6977	0.5703	0.5655
0.6678	3.0	72	0.6885	0.5652	0.5791
0.6458	4.0	96	0.6821	0.6162	0.5538
0.6044	5.0	120	0.6455	0.7632	0.6491
0.5292	6.0	144	0.6665	0.6913	0.5773
0.465	7.0	168	0.6464	0.7858	0.7436
0.405	8.0	192	0.6944	0.5644	0.4265
0.3644	9.0	216	0.6438	0.7717	0.5970
0.3322	10.0	240	0.5536	0.8986	0.7548
0.2943	11.0	264	0.6563	0.7126	0.5708
0.2759	12.0	288	0.6251	0.7273	0.6476
0.2611	13.0	312	0.4805	0.8944	0.8466
0.2564	14.0	336	0.5699	0.8439	0.7316
0.2295	15.0	360	0.5495	0.9470	0.5482
0.2248	16.0	384	0.5283	0.9119	0.7564
0.2236	17.0	408	0.6964	0.8571	0.5631
0.2213	18.0	432	0.4392	0.9343	0.8687
0.2137	19.0	456	0.5420	0.9090	0.7078
0.1924	20.0	480	0.4557	0.8704	0.7333
0.1886	21.0	504	0.3366	0.9336	0.8406
0.1698	22.0	528	0.6424	0.7500	0.5855
0.168	23.0	552	0.4096	0.9823	0.8052
0.1664	24.0	576	0.6902	0.7813	0.6339
0.1552	25.0	600	0.8798	0.7310	0.6081
0.1667	26.0	624	0.6585	0.6655	0.5795
0.1583	27.0	648	0.3810	0.9180	0.8713
0.1596	28.0	672	0.5435	0.8886	0.7953
0.1484	29.0	696	0.2422	0.9857	0.8907
0.1477	30.0	720	0.2892	0.9876	0.8804
0.1688	31.0	744	0.2391	0.9738	0.8930
0.1441	32.0	768	0.5261	0.8427	0.6695
0.1211	33.0	792	0.4854	0.8897	0.7593
0.1284	34.0	816	0.8883	0.9739	0.6192
0.1449	35.0	840	0.3004	0.9536	0.8748
0.1427	36.0	864	0.8280	0.8215	0.7558
0.1399	37.0	888	0.3803	0.9801	0.7871
0.1231	38.0	912	0.5545	0.9611	0.6467
0.1135	39.0	936	0.6676	0.9481	0.7822
0.1486	40.0	960	0.1412	0.9926	0.9471
0.1099	41.0	984	0.4054	0.9666	0.7790
0.1186	42.0	1008	0.4833	0.9629	0.8668
0.1246	43.0	1032	0.4189	0.9209	0.8191
0.1223	44.0	1056	0.5414	0.9904	0.7744
0.1073	45.0	1080	0.1448	0.9917	0.9424
0.1435	46.0	1104	0.5719	0.8858	0.7440
0.1102	47.0	1128	0.2645	0.9783	0.8805
0.0981	48.0	1152	0.8187	0.8929	0.7918
0.1141	49.0	1176	0.1664	0.9847	0.9341
0.1177	50.0	1200	0.1558	0.9975	0.9330
0.1121	51.0	1224	0.1854	0.9826	0.9203
0.1166	52.0	1248	0.1057	0.9968	0.9609
0.121	53.0	1272	0.1105	0.9947	0.9548
0.1232	54.0	1296	0.0675	0.9978	0.9739
0.1196	55.0	1320	0.0897	0.9964	0.9693
0.0961	56.0	1344	0.1438	0.9940	0.9401
0.1027	57.0	1368	0.0601	0.9986	0.9778
0.1175	58.0	1392	0.0943	0.9956	0.9647
0.1123	59.0	1416	0.0977	0.9945	0.9693
0.0999	60.0	1440	0.0712	0.9972	0.9739
0.1116	61.0	1464	0.0749	0.9970	0.9678
0.104	62.0	1488	0.0667	0.9983	0.9747
0.1051	63.0	1512	0.0653	0.9979	0.9762
0.1032	64.0	1536	0.0660	0.9978	0.9778
0.105	65.0	1560	0.0666	0.9979	0.9755
0.1066	66.0	1584	0.0635	0.9982	0.9778
0.1125	67.0	1608	0.0677	0.9979	0.9778
0.1077	68.0	1632	0.0603	0.9982	0.9801
0.0916	69.0	1656	0.0629	0.9980	0.9770
0.0985	70.0	1680	0.0610	0.9980	0.9824

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0