README.md

metadata

license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: INT03-PC
    results: []

INT03-PC

This model is a fine-tuned version of prajjwal1/bert-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0158
• Accuracy: 1.0
• F1: 1.0

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.0	50	0.6785	0.74	0.7413
No log	0.01	100	0.6047	0.79	0.7903
No log	0.01	150	0.4424	0.88	0.8769
No log	0.02	200	0.3601	0.89	0.8868
No log	0.02	250	0.3436	0.89	0.8868
No log	0.03	300	0.3311	0.9	0.8975
No log	0.03	350	0.3145	0.89	0.8876
No log	0.04	400	0.3113	0.9	0.8982
No log	0.04	450	0.2994	0.9	0.8982
0.4886	0.05	500	0.2806	0.88	0.8785
0.4886	0.05	550	0.2179	0.92	0.9194
0.4886	0.06	600	0.2388	0.88	0.8803
0.4886	0.06	650	0.1716	0.94	0.9398
0.4886	0.07	700	0.1774	0.93	0.9302
0.4886	0.07	750	0.1456	0.95	0.9501
0.4886	0.08	800	0.1518	0.94	0.9402
0.4886	0.08	850	0.1564	0.93	0.9303
0.4886	0.09	900	0.1684	0.92	0.9204
0.4886	0.09	950	0.1372	0.96	0.9602
0.2446	0.1	1000	0.1368	0.94	0.9402
0.2446	0.1	1050	0.1502	0.95	0.9502
0.2446	0.11	1100	0.1385	0.95	0.9502
0.2446	0.11	1150	0.1297	0.96	0.9602
0.2446	0.12	1200	0.1917	0.95	0.9502
0.2446	0.12	1250	0.1042	0.97	0.9700
0.2446	0.13	1300	0.1502	0.96	0.9602
0.2446	0.13	1350	0.1436	0.96	0.9602
0.2446	0.14	1400	0.0896	0.98	0.9800
0.2446	0.14	1450	0.1045	0.96	0.9602
0.1824	0.15	1500	0.1269	0.96	0.9602
0.1824	0.15	1550	0.1449	0.96	0.9602
0.1824	0.16	1600	0.1311	0.96	0.9602
0.1824	0.16	1650	0.1380	0.96	0.9602
0.1824	0.17	1700	0.1466	0.96	0.9602
0.1824	0.17	1750	0.0861	0.98	0.9800
0.1824	0.18	1800	0.1323	0.96	0.9602
0.1824	0.18	1850	0.1375	0.96	0.9602
0.1824	0.19	1900	0.1719	0.96	0.9602
0.1824	0.19	1950	0.0837	0.98	0.9800
0.1252	0.2	2000	0.1661	0.96	0.9602
0.1252	0.2	2050	0.1129	0.96	0.9602
0.1252	0.21	2100	0.0603	0.98	0.9800
0.1252	0.21	2150	0.1231	0.96	0.9602
0.1252	0.22	2200	0.1363	0.96	0.9602
0.1252	0.22	2250	0.1330	0.96	0.9602
0.1252	0.23	2300	0.0795	0.96	0.9602
0.1252	0.23	2350	0.0856	0.96	0.9602

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0