finetuning-sentiment

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: 0.8125
• Accuracy@en: 0.9033
• F1@en: 0.9002
• Precision@en: 0.8989
• Recall@en: 0.9018
• Loss@en: 0.8125

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: 8
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 13

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy@en	F1@en	Precision@en	Recall@en	Loss@en
No log	1.0	375	0.4653	0.8933	0.8895	0.8895	0.8895	0.4653
0.2086	2.0	750	0.4367	0.9033	0.9011	0.8979	0.9069	0.4367
0.1622	3.0	1125	0.4866	0.91	0.9081	0.9047	0.9151	0.4866
0.0622	4.0	1500	0.6156	0.9	0.8982	0.8951	0.9067	0.6156
0.0622	5.0	1875	0.6790	0.9133	0.9102	0.9102	0.9102	0.6790
0.0193	6.0	2250	0.6822	0.9	0.8978	0.8945	0.9041	0.6822
0.0202	7.0	2625	0.6595	0.91	0.9077	0.9047	0.9126	0.6595
0.0148	8.0	3000	0.6538	0.9067	0.9042	0.9014	0.9085	0.6538
0.0148	9.0	3375	0.6869	0.9067	0.9050	0.9018	0.9136	0.6869
0.0036	10.0	3750	0.7016	0.9033	0.9007	0.8981	0.9044	0.7016
0.0038	11.0	4125	0.8170	0.9	0.8961	0.8972	0.8951	0.8170
0.008	12.0	4500	0.8169	0.9033	0.9002	0.8989	0.9018	0.8169
0.008	13.0	4875	0.8125	0.9033	0.9002	0.8989	0.9018	0.8125

Framework versions

• Transformers 4.17.0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2