README.md

metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
model-index:
  - name: distilbert-base-uncased-finetuned-pfe-projectt
    results: []

distilbert-base-uncased-finetuned-pfe-projectt

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

• Loss: 1.5126
• Start Accuracy: 0.7021
• End Accuracy: 0.6241
• Overall Accuracy: 0.6631
• Start Precision: 0.2862
• End Precision: 0.1446
• Start Recall: 0.3165
• End Recall: 0.1961
• Start F1 Score: 0.2802
• End F1 Score: 0.1506

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 0
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Start Accuracy	End Accuracy	Overall Accuracy	Start Precision	End Precision	Start Recall	End Recall	Start F1 Score	End F1 Score
4.8235	1.0	17	3.6096	0.4752	0.4752	0.4752	0.0153	0.0140	0.0323	0.0294	0.0208	0.0189
2.5596	2.0	34	2.5775	0.4752	0.4752	0.4752	0.0153	0.0140	0.0323	0.0294	0.0208	0.0189
2.0949	3.0	51	2.2935	0.4752	0.4752	0.4752	0.0153	0.0140	0.0323	0.0294	0.0208	0.0189
1.9248	4.0	68	2.1906	0.4752	0.4752	0.4752	0.0153	0.0140	0.0323	0.0294	0.0208	0.0189
1.7309	5.0	85	2.0463	0.5674	0.4752	0.5213	0.0978	0.0140	0.1340	0.0294	0.1110	0.0189
1.5193	6.0	102	1.9047	0.5745	0.4823	0.5284	0.1116	0.0390	0.1544	0.0657	0.1240	0.0438
1.4401	7.0	119	1.8648	0.5532	0.5035	0.5284	0.0960	0.0686	0.1271	0.0858	0.1046	0.0676
1.3916	8.0	136	1.7904	0.6170	0.5532	0.5851	0.1673	0.0709	0.2090	0.1041	0.1800	0.0795
1.2498	9.0	153	1.8084	0.6170	0.5816	0.5993	0.1309	0.0862	0.1791	0.1145	0.1449	0.0930
1.1733	10.0	170	1.7518	0.6241	0.5957	0.6099	0.1586	0.0925	0.2032	0.1221	0.1712	0.0995
1.0563	11.0	187	1.6420	0.6241	0.5816	0.6028	0.1673	0.0918	0.2206	0.1337	0.1814	0.1047
1.0074	12.0	204	1.8142	0.6454	0.6099	0.6277	0.1453	0.0934	0.1737	0.1080	0.1497	0.0982
0.9473	13.0	221	1.6035	0.6738	0.6241	0.6489	0.2440	0.1010	0.2606	0.1444	0.2340	0.1138
0.9307	14.0	238	1.4999	0.6809	0.6241	0.6525	0.2226	0.1151	0.2484	0.1418	0.2225	0.1141
0.8668	15.0	255	1.5837	0.6950	0.6312	0.6631	0.2456	0.1033	0.2678	0.1520	0.2395	0.1162
0.8226	16.0	272	1.5517	0.6879	0.6312	0.6596	0.2741	0.1385	0.2911	0.1955	0.2523	0.1451
0.7358	17.0	289	1.5387	0.7092	0.6241	0.6667	0.3022	0.1374	0.3360	0.1898	0.2854	0.1424
0.7529	18.0	306	1.4644	0.6950	0.6383	0.6667	0.2554	0.1358	0.3011	0.1947	0.2531	0.1450
0.6962	19.0	323	1.5374	0.6809	0.6383	0.6596	0.2570	0.1419	0.2890	0.1861	0.2516	0.1502
0.6807	20.0	340	1.4873	0.6809	0.6383	0.6596	0.2577	0.1469	0.3058	0.1840	0.2623	0.1572
0.6988	21.0	357	1.5178	0.6667	0.6099	0.6383	0.2843	0.1558	0.3050	0.1982	0.2802	0.1597
0.6632	22.0	374	1.5319	0.7092	0.6312	0.6702	0.2860	0.1489	0.3053	0.2015	0.2788	0.1628
0.6081	23.0	391	1.5817	0.7021	0.6454	0.6738	0.2537	0.1767	0.2853	0.1944	0.2489	0.1677
0.5926	24.0	408	1.5514	0.6950	0.6241	0.6596	0.3069	0.1451	0.3160	0.1913	0.2937	0.1465
0.6449	25.0	425	1.5521	0.6950	0.6383	0.6667	0.2826	0.1687	0.3336	0.2133	0.2806	0.1654
0.5542	26.0	442	1.4860	0.7305	0.6383	0.6844	0.3305	0.1598	0.3734	0.1812	0.3257	0.1580
0.5668	27.0	459	1.5091	0.7092	0.6454	0.6773	0.3201	0.1688	0.3304	0.2075	0.2982	0.1670
0.5603	28.0	476	1.5088	0.7021	0.6454	0.6738	0.2819	0.1535	0.3165	0.2022	0.2772	0.1569
0.5353	29.0	493	1.5101	0.7021	0.6383	0.6702	0.2862	0.1447	0.3165	0.1968	0.2802	0.1494
0.5082	30.0	510	1.5126	0.7021	0.6241	0.6631	0.2862	0.1446	0.3165	0.1961	0.2802	0.1506

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2