Demo

👉 Launch Gradio Demo 🚀

DistilBERT_IMDB

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: 0.2407
• Accuracy: 0.9156
• F1: 0.9153

Model description

• Base Model - Distilbert-base-uncased
• Fine-tuned for binary classification
• Achieved ~90% accuracy on test set

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.06
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.5779	0.16	100	0.5232	0.8628	0.8565
0.3239	0.32	200	0.3461	0.8596	0.8408
0.2367	0.48	300	0.2935	0.8806	0.8863
0.2037	0.64	400	0.2547	0.9006	0.8968
0.2215	0.8	500	0.2354	0.908	0.9064
0.1866	0.96	600	0.2462	0.9046	0.9063
0.161	1.12	700	0.2435	0.911	0.9095
0.2101	1.28	800	0.2407	0.9156	0.9153

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0