|
--- |
|
license: mit |
|
base_model: intfloat/multilingual-e5-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: multi-e5-base_lmd-comments_v1 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# multi-e5-base_lmd-comments_v1 |
|
|
|
This model is a fine-tuned version of [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.1145 |
|
- F1: 0.7338 |
|
- Accuracy: 0.7410 |
|
|
|
## 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: 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_ratio: 0.1 |
|
- num_epochs: 2 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:| |
|
| 1.1179 | 0.04 | 100 | 1.1781 | 0.3969 | 0.4748 | |
|
| 1.0342 | 0.08 | 200 | 1.0924 | 0.5137 | 0.5899 | |
|
| 0.7423 | 0.12 | 300 | 0.9700 | 0.6454 | 0.6691 | |
|
| 0.7046 | 0.17 | 400 | 0.8990 | 0.6462 | 0.6691 | |
|
| 0.6419 | 0.21 | 500 | 0.9583 | 0.6220 | 0.6475 | |
|
| 0.6679 | 0.25 | 600 | 0.8621 | 0.6757 | 0.6835 | |
|
| 0.6244 | 0.29 | 700 | 0.8017 | 0.7399 | 0.7410 | |
|
| 0.5747 | 0.33 | 800 | 0.8040 | 0.6950 | 0.6906 | |
|
| 0.5575 | 0.37 | 900 | 1.1045 | 0.6774 | 0.6906 | |
|
| 0.5994 | 0.41 | 1000 | 1.1592 | 0.6725 | 0.6978 | |
|
| 0.5565 | 0.46 | 1100 | 0.9960 | 0.7303 | 0.7338 | |
|
| 0.511 | 0.5 | 1200 | 1.0861 | 0.7377 | 0.7482 | |
|
| 0.5448 | 0.54 | 1300 | 0.7945 | 0.7155 | 0.7122 | |
|
| 0.6059 | 0.58 | 1400 | 0.8167 | 0.6879 | 0.6906 | |
|
| 0.4865 | 0.62 | 1500 | 1.1002 | 0.7181 | 0.7266 | |
|
| 0.566 | 0.66 | 1600 | 0.7388 | 0.6678 | 0.6691 | |
|
| 0.4756 | 0.7 | 1700 | 1.1751 | 0.7385 | 0.7482 | |
|
| 0.5595 | 0.75 | 1800 | 1.0169 | 0.7204 | 0.7266 | |
|
| 0.5838 | 0.79 | 1900 | 0.7718 | 0.7005 | 0.6978 | |
|
| 0.573 | 0.83 | 2000 | 0.9156 | 0.7174 | 0.7266 | |
|
| 0.5623 | 0.87 | 2100 | 0.8405 | 0.7416 | 0.7482 | |
|
| 0.4929 | 0.91 | 2200 | 0.8329 | 0.7484 | 0.7554 | |
|
| 0.5135 | 0.95 | 2300 | 1.1845 | 0.7008 | 0.7194 | |
|
| 0.5217 | 0.99 | 2400 | 1.1482 | 0.7204 | 0.7338 | |
|
| 0.4342 | 1.04 | 2500 | 1.3326 | 0.7078 | 0.7266 | |
|
| 0.4975 | 1.08 | 2600 | 1.0527 | 0.7048 | 0.7194 | |
|
| 0.4135 | 1.12 | 2700 | 0.9742 | 0.7431 | 0.7482 | |
|
| 0.3562 | 1.16 | 2800 | 1.0554 | 0.7359 | 0.7410 | |
|
| 0.3892 | 1.2 | 2900 | 1.1289 | 0.7403 | 0.7482 | |
|
| 0.5041 | 1.24 | 3000 | 0.9690 | 0.7642 | 0.7698 | |
|
| 0.4808 | 1.28 | 3100 | 0.9745 | 0.7378 | 0.7410 | |
|
| 0.3532 | 1.33 | 3200 | 1.0141 | 0.7521 | 0.7554 | |
|
| 0.4679 | 1.37 | 3300 | 0.9923 | 0.7410 | 0.7482 | |
|
| 0.432 | 1.41 | 3400 | 1.0650 | 0.7486 | 0.7554 | |
|
| 0.4543 | 1.45 | 3500 | 1.1235 | 0.7474 | 0.7554 | |
|
| 0.4716 | 1.49 | 3600 | 1.0688 | 0.7316 | 0.7410 | |
|
| 0.4251 | 1.53 | 3700 | 1.0290 | 0.7415 | 0.7482 | |
|
| 0.3676 | 1.57 | 3800 | 1.1651 | 0.7546 | 0.7626 | |
|
| 0.4031 | 1.62 | 3900 | 0.9981 | 0.7559 | 0.7626 | |
|
| 0.4356 | 1.66 | 4000 | 0.9815 | 0.7558 | 0.7626 | |
|
| 0.4355 | 1.7 | 4100 | 1.0349 | 0.7443 | 0.7482 | |
|
| 0.4113 | 1.74 | 4200 | 1.1226 | 0.7333 | 0.7410 | |
|
| 0.4447 | 1.78 | 4300 | 0.9854 | 0.7423 | 0.7482 | |
|
| 0.4601 | 1.82 | 4400 | 1.0193 | 0.7348 | 0.7410 | |
|
| 0.4474 | 1.86 | 4500 | 1.0177 | 0.7423 | 0.7482 | |
|
| 0.3585 | 1.91 | 4600 | 1.0460 | 0.7276 | 0.7338 | |
|
| 0.4064 | 1.95 | 4700 | 1.0995 | 0.7276 | 0.7338 | |
|
| 0.4443 | 1.99 | 4800 | 1.1145 | 0.7338 | 0.7410 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.1 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.15.2 |
|
|
