metadata
library_name: transformers
license: mit
base_model: BAAI/bge-small-en-v1.5
tags:
- generated_from_trainer
model-index:
- name: bge-small-en-v1.5-2024-12-08_01-40-13-quality-weight-0.5
results: []
bge-small-en-v1.5-2024-12-08_01-40-13-quality-weight-0.5
This model is a fine-tuned version of BAAI/bge-small-en-v1.5 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0194
- Spearman: 0.9296
- Pearson: 0.9282
- Mse: 0.0194
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: 64
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Spearman | Pearson | Mse |
|---|---|---|---|---|---|---|
| 0.0295 | 0.3998 | 1055 | 0.0259 | 0.9011 | 0.9003 | 0.0259 |
| 0.0247 | 0.7997 | 2110 | 0.0233 | 0.9112 | 0.9111 | 0.0233 |
| 0.0218 | 1.1995 | 3165 | 0.0227 | 0.9170 | 0.9169 | 0.0227 |
| 0.0213 | 1.5994 | 4220 | 0.0204 | 0.9226 | 0.9222 | 0.0204 |
| 0.019 | 1.9992 | 5275 | 0.0206 | 0.9240 | 0.9245 | 0.0206 |
| 0.0168 | 2.3991 | 6330 | 0.0199 | 0.9248 | 0.9265 | 0.0199 |
| 0.0155 | 2.7989 | 7385 | 0.0193 | 0.9268 | 0.9281 | 0.0193 |
| 0.0123 | 3.1988 | 8440 | 0.0195 | 0.9278 | 0.9273 | 0.0195 |
| 0.0134 | 3.5986 | 9495 | 0.0192 | 0.9289 | 0.9293 | 0.0192 |
| 0.0136 | 3.9985 | 10550 | 0.0190 | 0.9290 | 0.9296 | 0.0190 |
| 0.0103 | 4.3983 | 11605 | 0.0193 | 0.9291 | 0.9294 | 0.0193 |
| 0.0117 | 4.7982 | 12660 | 0.0193 | 0.9293 | 0.9295 | 0.0193 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.1+cu121
- Datasets 2.19.2
- Tokenizers 0.20.3