Edit model card

microsoft-codebert-base-finetuned-defect-detection

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

  • Loss: 0.6197
  • Accuracy: 0.7382
  • Roc Auc: 0.7394
  • Precision: 0.7070
  • Recall: 0.7924

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: 8
  • eval_batch_size: 8
  • seed: 4711
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Roc Auc Precision Recall
0.6456 1.0 996 0.5435 0.6832 0.6810 0.7151 0.5843
0.5086 2.0 1993 0.5373 0.7113 0.7139 0.6654 0.8227
0.4173 3.0 2989 0.5476 0.7289 0.7293 0.7125 0.7461
0.3543 4.0 3986 0.5803 0.7357 0.7369 0.7051 0.7888
0.3059 5.0 4980 0.6197 0.7382 0.7394 0.7070 0.7924

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.2.0+cu121
  • Datasets 2.17.1
  • Tokenizers 0.15.2
Downloads last month
22
Safetensors
Model size
125M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for mcanoglu/microsoft-codebert-base-finetuned-defect-detection

Finetuned
(24)
this model