Edit model card

albert-base-ours-run-1

This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3970
  • Accuracy: 0.735
  • Precision: 0.7033
  • Recall: 0.6790
  • F1: 0.6873

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: 1e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • 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 Precision Recall F1
0.9719 1.0 200 0.8460 0.635 0.6534 0.5920 0.5547
0.7793 2.0 400 0.7762 0.675 0.6965 0.6323 0.5936
0.5734 3.0 600 0.8149 0.67 0.6200 0.6192 0.6196
0.3877 4.0 800 0.9555 0.7 0.6724 0.6482 0.6549
0.2426 5.0 1000 1.1248 0.695 0.6529 0.6437 0.6452
0.183 6.0 1200 1.3497 0.705 0.6717 0.6489 0.6563
0.1011 7.0 1400 1.6369 0.7 0.6620 0.6532 0.6560
0.0602 8.0 1600 1.8171 0.7 0.6763 0.6615 0.6654
0.0335 9.0 1800 1.9601 0.695 0.6640 0.6490 0.6545
0.0158 10.0 2000 2.0206 0.71 0.6802 0.6751 0.6768
0.0148 11.0 2200 2.0881 0.675 0.6252 0.6242 0.6232
0.0057 12.0 2400 2.2708 0.735 0.7146 0.6790 0.6904
0.0079 13.0 2600 2.2348 0.72 0.6917 0.6659 0.6746
0.0018 14.0 2800 2.2978 0.725 0.6968 0.6662 0.6761
0.0025 15.0 3000 2.3180 0.735 0.7067 0.6790 0.6883
0.0028 16.0 3200 2.3910 0.74 0.7153 0.6854 0.6953
0.0002 17.0 3400 2.3830 0.735 0.7033 0.6790 0.6873
0.0002 18.0 3600 2.3899 0.735 0.7033 0.6790 0.6873
0.0001 19.0 3800 2.3922 0.735 0.7033 0.6790 0.6873
0.0001 20.0 4000 2.3970 0.735 0.7033 0.6790 0.6873

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.0+cu116
  • Tokenizers 0.13.2
Downloads last month
10
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 SkyR/albert-base-ours-run-1

Finetuned
(162)
this model