Edit model card

e5_finetuned

This model is a fine-tuned version of intfloat/multilingual-e5-small on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0611
  • Precision: 0.9494
  • Recall: 0.8860
  • F1: 0.9166
  • Accuracy: 0.9799

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
  • 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
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 5.0

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 0.0009 2 0.7141 0.125 1.0 0.2222 0.125
0.1046 0.9998 2334 0.0905 0.9564 0.8239 0.8852 0.9733
0.0786 2.0 4669 0.0734 0.9550 0.8540 0.9016 0.9767
0.0761 2.9998 7003 0.0690 0.9358 0.8834 0.9088 0.9778
0.0673 4.0 9338 0.0621 0.9594 0.8750 0.9152 0.9797
0.0709 4.9989 11670 0.0611 0.9494 0.8860 0.9166 0.9799

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
5
Safetensors
Model size
118M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for wl-tookitaki/e5_finetuned

Finetuned
(56)
this model