README.md

metadata

base_model: uitnlp/visobert
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: VP_ViSoBERT_syl_ViWikiFC
    results: []

VP_ViSoBERT_syl_ViWikiFC

This model is a fine-tuned version of uitnlp/visobert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9243
• Accuracy: 0.6364

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: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1299	0.1	100	1.1182	0.3411
1.0816	0.19	200	1.0678	0.3976
1.0181	0.29	300	1.0163	0.4823
1.0121	0.38	400	0.9956	0.5072
0.9617	0.48	500	0.9718	0.5048
0.9297	0.57	600	0.9665	0.5239
0.9332	0.67	700	0.9252	0.5646
0.9057	0.76	800	0.9667	0.5421
0.8756	0.86	900	0.8884	0.5871
0.879	0.96	1000	0.8907	0.5718
0.8249	1.05	1100	0.8793	0.5981
0.7177	1.15	1200	0.8951	0.5957
0.7145	1.24	1300	0.9523	0.6062
0.7469	1.34	1400	0.9001	0.5986
0.7358	1.43	1500	0.8865	0.6081
0.7112	1.53	1600	0.9099	0.6057
0.7299	1.62	1700	0.8496	0.6144
0.6949	1.72	1800	0.8580	0.6124
0.6988	1.81	1900	0.8840	0.6215
0.6524	1.91	2000	0.8753	0.6134
0.6914	2.01	2100	0.8729	0.6330
0.5427	2.1	2200	0.9494	0.6431
0.5628	2.2	2300	0.9531	0.6120
0.5607	2.29	2400	0.9050	0.6340
0.5396	2.39	2500	0.9149	0.6335
0.5178	2.48	2600	0.9848	0.6124
0.5322	2.58	2700	0.9198	0.6330
0.5406	2.67	2800	0.9206	0.6364
0.5183	2.77	2900	0.9150	0.6392
0.5369	2.87	3000	0.9200	0.6340
0.5105	2.96	3100	0.9243	0.6364

Framework versions

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