2025-24679-HW2-Part3-text-distilbert-predictor

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0047
• Accuracy: 1.0
• F1: 1.0
• Precision: 1.0
• Recall: 1.0

Model description

This model classifies short text descriptions as either representing East Coast or West Coast culture.
It was fine-tuned on a small dataset of regionally themed examples.

Intended uses & limitations

• Intended use: educational purposes, small-scale demos of text classification with Hugging Face transformers.
• Limitations: the dataset is small and subjective, so predictions may not generalize beyond the training examples.

Training and evaluation data

The dataset consisted of short text snippets describing lifestyle, weather, food, and culture, each labeled as East Coast (0) or West Coast (1).

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.0064	1.0	120	0.0052	1.0	1.0	1.0	1.0
0.0016	2.0	240	0.0039	1.0	1.0	1.0	1.0
0.0009	3.0	360	0.0007	1.0	1.0	1.0	1.0
0.0006	4.0	480	0.0005	1.0	1.0	1.0	1.0
0.0006	5.0	600	0.0005	1.0	1.0	1.0	1.0

Framework versions

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