output_LiLT_test_03

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

  • Loss: 0.1452
  • Precision: 0.7549
  • Recall: 0.8276
  • F1: 0.7896
  • Accuracy: 0.9577

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: 5e-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
  • lr_scheduler_warmup_steps: 300
  • num_epochs: 45

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 0.0808 100 0.5589 0.0259 0.0020 0.0037 0.8724
No log 0.1617 200 0.3198 0.3704 0.3698 0.3701 0.9009
No log 0.2425 300 0.2222 0.6005 0.6743 0.6353 0.9346
No log 0.3234 400 0.1950 0.7231 0.6316 0.6743 0.9446
0.4262 0.4042 500 0.1684 0.7115 0.7215 0.7165 0.9491
0.4262 0.4850 600 0.1885 0.6328 0.7602 0.6907 0.9366
0.4262 0.5659 700 0.1776 0.6609 0.8097 0.7278 0.9437
0.4262 0.6467 800 0.2086 0.6420 0.8034 0.7137 0.9398
0.4262 0.7276 900 0.1702 0.8241 0.7469 0.7836 0.9611
0.1027 0.8084 1000 0.1496 0.8028 0.7558 0.7786 0.9601
0.1027 0.8892 1100 0.1620 0.7354 0.7754 0.7549 0.9513
0.1027 0.9701 1200 0.1404 0.8062 0.7997 0.8029 0.9627
0.1027 1.0509 1300 0.1626 0.7976 0.7564 0.7764 0.9597
0.1027 1.1318 1400 0.1741 0.7253 0.8089 0.7648 0.9485
0.0764 1.2126 1500 0.1278 0.8170 0.7798 0.7979 0.9645
0.0764 1.2935 1600 0.1452 0.7549 0.8276 0.7896 0.9577

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
11
Safetensors
Model size
284M 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 benedma2/output_LiLT_test_03

Finetuned
(29)
this model