multibert_seed35_1311

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

• Loss: 0.4565
• Precisions: 0.8851
• Recall: 0.8156
• F-measure: 0.8447
• Accuracy: 0.9385

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: 7.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 35
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Precisions	Recall	F-measure	Accuracy
0.4595	1.0	236	0.2631	0.8847	0.7209	0.7552	0.9209
0.2301	2.0	472	0.2468	0.8447	0.7729	0.7954	0.9224
0.1332	3.0	708	0.2457	0.8468	0.7925	0.8086	0.9301
0.082	4.0	944	0.3261	0.8749	0.7855	0.8184	0.9303
0.0511	5.0	1180	0.3209	0.8952	0.8048	0.8379	0.9405
0.0363	6.0	1416	0.3568	0.8594	0.8056	0.8290	0.9337
0.0214	7.0	1652	0.3562	0.8851	0.8068	0.8350	0.9395
0.0152	8.0	1888	0.4160	0.8679	0.8065	0.8309	0.9371
0.0136	9.0	2124	0.4247	0.8732	0.8103	0.8375	0.9342
0.0096	10.0	2360	0.4242	0.8864	0.8041	0.8381	0.9378
0.0036	11.0	2596	0.4306	0.8746	0.8122	0.8365	0.9373
0.0029	12.0	2832	0.4420	0.8744	0.8220	0.8446	0.9385
0.0014	13.0	3068	0.4526	0.8850	0.8090	0.8395	0.9376
0.002	14.0	3304	0.4565	0.8851	0.8156	0.8447	0.9385

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1