finetuning-sentiment-model-distil-samples

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

• Loss: 1.3150
• Accuracy Percentage: 0.7514
• Accuracy Number: 133.0
• F1: 0.7460
• Precision: 0.7514
• Recall: 0.7514

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Percentage	Accuracy Number	F1	Precision	Recall
0.2349	1.0	22	0.6664	0.7571	134.0	0.7552	0.7571	0.7571
0.0531	2.0	44	1.0491	0.7232	128.0	0.7093	0.7232	0.7232
0.0374	3.0	66	1.1389	0.7119	126.0	0.7154	0.7119	0.7119
0.023	4.0	88	1.2514	0.7401	131.0	0.7288	0.7401	0.7401
0.0188	5.0	110	1.2064	0.7401	131.0	0.7355	0.7401	0.7401
0.0171	6.0	132	1.3531	0.7458	132.0	0.7365	0.7458	0.7458
0.0188	7.0	154	1.3221	0.7627	135.0	0.7534	0.7627	0.7627
0.0162	8.0	176	1.2874	0.7571	134.0	0.7507	0.7571	0.7571
0.018	9.0	198	1.2882	0.7627	135.0	0.7579	0.7627	0.7627
0.0097	10.0	220	1.3150	0.7514	133.0	0.7460	0.7514	0.7514

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3