my-finetuned-FinanceNews-distilbert

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

• Loss: 0.3800

• Accuracy: 0.8557

• F1: 0.8547

• Precision: 0.8548

• Recall: 0.8557

• Classification Report: precision recall f1-score support

  Class 0 0.87 0.90 0.88 87 Class 1 0.87 0.90 0.88 268 Class 2 0.83 0.76 0.79 151

  accuracy 0.86 506 macro avg 0.85 0.85 0.85 506

weighted avg 0.85 0.86 0.85 506

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Classification Report
0.7555	1.0	72	0.4573	0.8241	0.8235	0.8232	0.8241	precision recall f1-score support

 Class 0       0.83      0.84      0.83        87
 Class 1       0.85      0.87      0.86       268
 Class 2       0.78      0.74      0.76       151

accuracy                           0.82       506

macro avg 0.82 0.82 0.82 506 weighted avg 0.82 0.82 0.82 506 | | 0.4201 | 2.0 | 144 | 0.3800 | 0.8557 | 0.8547 | 0.8548 | 0.8557 | precision recall f1-score support

 Class 0       0.87      0.90      0.88        87
 Class 1       0.87      0.90      0.88       268
 Class 2       0.83      0.76      0.79       151

accuracy                           0.86       506

macro avg 0.85 0.85 0.85 506 weighted avg 0.85 0.86 0.85 506 |

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.2
• Datasets 2.20.0
• Tokenizers 0.19.1