BotnoiNLPteam/aunt_mai_ork_v2

wisesight_sentiment

This model is a fine-tuned version of airesearch/wangchanberta-base-att-spm-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4140
• Accuracy: 0.9025
• F1 Micro: 0.9025
• Precision Micro: 0.9025
• Recall Micro: 0.9025
• F1 Macro: 0.9010
• Precision Macro: 0.9006
• Recall Macro: 0.9015
• Nb Samples: 1015

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 1412
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2031
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Micro	Precision Micro	Recall Micro	F1 Macro	Precision Macro	Recall Macro	Nb Samples
0.587	1.0	2031	0.5865	0.8532	0.8532	0.8532	0.8532	0.8492	0.8550	0.8460	1015
0.5212	2.0	4062	0.7026	0.8158	0.8158	0.8158	0.8158	0.8156	0.8380	0.8322	1015
0.4554	3.0	6093	0.4360	0.8768	0.8768	0.8768	0.8768	0.8731	0.8813	0.8690	1015
0.3752	4.0	8124	0.4121	0.8897	0.8897	0.8897	0.8897	0.8867	0.8929	0.8831	1015
0.382	5.0	10155	0.4709	0.8916	0.8916	0.8916	0.8916	0.8894	0.8917	0.8876	1015
0.3465	6.0	12186	0.3929	0.8975	0.8975	0.8975	0.8975	0.8957	0.8964	0.8952	1015
0.3246	7.0	14217	0.4801	0.8897	0.8897	0.8897	0.8897	0.8873	0.8899	0.8854	1015
0.2808	8.0	16248	0.5070	0.8897	0.8897	0.8897	0.8897	0.8881	0.8874	0.8889	1015
0.2786	9.0	18279	0.4098	0.8995	0.8995	0.8995	0.8995	0.8975	0.8993	0.8962	1015
0.2151	10.0	20310	0.4140	0.9025	0.9025	0.9025	0.9025	0.9010	0.9006	0.9015	1015

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1