2025-24679-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: 0.0451
• Accuracy: 1.0
• F1: 1.0
• Precision: 1.0
• Recall: 1.0

Model description

Purpose: This model to be used for in-class assignments and activity associated with Course 24679 at CMU.

Preprocessing/Augmentation: The preprocessing of this data includes splitting the dataset into train and test, and using autoML to predict whether the book from the dataset will be reccomended The peredictor weas fitting using a 20 minute time limit, in addition to a best_quality present and auto_stacking to improve accuracy over the constrained timeframe.

Intended uses & limitations

Intented use/limits: The intended use of this dataset is exclusively for classroom and assignment use. Please request permission if you wish to use it elsewhere

Ethical notes: The AI used in this sample is a fairly benign use case handling basic text manipulation. However, please review the environmental impacts of large-scale usage in exchange for implementing the necessary good such technology brings.

AI Usage disclosure: Original code to assist with data augmentation was developed with use of Google Gemini in combination with course material from 24679 at CMU

Training and evaluation data

limitations:

Dataset trained on is a simple text dataset for books, and has been trained to classify the reccomended column which is a simple binary. This training model is quite minimal as a result

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
0.0527	1.0	80	0.0426	1.0	1.0	1.0	1.0
0.0088	2.0	160	0.0070	1.0	1.0	1.0	1.0
0.006	3.0	240	0.0041	1.0	1.0	1.0	1.0
0.0042	4.0	320	0.0032	1.0	1.0	1.0	1.0
0.0042	5.0	400	0.0029	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