model_ckpt

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: 0.0091
  • Accuracy: 0.9989
  • F1: 0.9989
  • Precision: 0.9989
  • Recall: 0.9989

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: 0.0001
  • train_batch_size: 116
  • eval_batch_size: 116
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.3936 1.0 45 0.0168 1.0 1.0 1.0 1.0
0.0187 2.0 90 0.0047 0.9989 0.9989 0.9989 0.9989
0.0049 3.0 135 0.0345 0.9945 0.9945 0.9945 0.9945
0.0035 4.0 180 0.0105 0.9978 0.9978 0.9978 0.9978
0.0015 5.0 225 0.0075 0.9989 0.9989 0.9989 0.9989
0.0009 6.0 270 0.0077 0.9989 0.9989 0.9989 0.9989
0.0007 7.0 315 0.0078 0.9989 0.9989 0.9989 0.9989
0.0006 8.0 360 0.0079 0.9989 0.9989 0.9989 0.9989
0.0005 9.0 405 0.0081 0.9989 0.9989 0.9989 0.9989
0.0004 10.0 450 0.0083 0.9989 0.9989 0.9989 0.9989
0.0004 11.0 495 0.0085 0.9989 0.9989 0.9989 0.9989
0.0003 12.0 540 0.0086 0.9989 0.9989 0.9989 0.9989
0.0003 13.0 585 0.0087 0.9989 0.9989 0.9989 0.9989
0.0003 14.0 630 0.0088 0.9989 0.9989 0.9989 0.9989
0.0003 15.0 675 0.0089 0.9989 0.9989 0.9989 0.9989
0.0003 16.0 720 0.0090 0.9989 0.9989 0.9989 0.9989
0.0002 17.0 765 0.0090 0.9989 0.9989 0.9989 0.9989
0.0002 18.0 810 0.0090 0.9989 0.9989 0.9989 0.9989
0.0002 19.0 855 0.0091 0.9989 0.9989 0.9989 0.9989
0.0002 20.0 900 0.0091 0.9989 0.9989 0.9989 0.9989

Framework versions

  • Transformers 4.42.3
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
56
Safetensors
Model size
67M 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 RugvedaWork/model_ckpt

Finetuned
(7076)
this model