metadata
license: apache-2.0
base_model: shibing624/text2vec-base-chinese
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: cosent-similarity-text2vec
results: []
cosent-similarity-text2vec
This model is a fine-tuned version of shibing624/text2vec-base-chinese on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1408
- Accuracy: 0.9605
- F1: 0.9670
- Precision: 0.9778
- Recall: 0.9565
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 22 | 0.2330 | 0.9211 | 0.9318 | 0.9762 | 0.8913 |
| No log | 2.0 | 44 | 0.2088 | 0.9342 | 0.9438 | 0.9767 | 0.9130 |
| No log | 3.0 | 66 | 0.1484 | 0.9605 | 0.9670 | 0.9778 | 0.9565 |
| No log | 4.0 | 88 | 0.1370 | 0.9605 | 0.9670 | 0.9778 | 0.9565 |
| No log | 5.0 | 110 | 0.1408 | 0.9605 | 0.9670 | 0.9778 | 0.9565 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.1