|
--- |
|
base_model: microsoft/mdeberta-v3-base |
|
library_name: transformers |
|
license: mit |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: scenario-kd-pre-ner-full-mdeberta_data-univner_full44 |
|
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. --> |
|
|
|
# scenario-kd-pre-ner-full-mdeberta_data-univner_full44 |
|
|
|
This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2650 |
|
- Precision: 0.8107 |
|
- Recall: 0.8117 |
|
- F1: 0.8112 |
|
- Accuracy: 0.9806 |
|
|
|
## 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: 3e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 32 |
|
- seed: 44 |
|
- 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 |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 1.3559 | 0.2911 | 500 | 0.7891 | 0.4246 | 0.3525 | 0.3852 | 0.9433 | |
|
| 0.7017 | 0.5822 | 1000 | 0.5422 | 0.6316 | 0.6298 | 0.6307 | 0.9646 | |
|
| 0.528 | 0.8732 | 1500 | 0.4693 | 0.6856 | 0.6830 | 0.6843 | 0.9692 | |
|
| 0.4354 | 1.1643 | 2000 | 0.4211 | 0.7101 | 0.7376 | 0.7236 | 0.9724 | |
|
| 0.385 | 1.4554 | 2500 | 0.3893 | 0.7482 | 0.7374 | 0.7428 | 0.9747 | |
|
| 0.3575 | 1.7465 | 3000 | 0.3713 | 0.7678 | 0.7331 | 0.7500 | 0.9752 | |
|
| 0.3298 | 2.0375 | 3500 | 0.3550 | 0.7497 | 0.7800 | 0.7645 | 0.9761 | |
|
| 0.2879 | 2.3286 | 4000 | 0.3492 | 0.7964 | 0.7367 | 0.7654 | 0.9763 | |
|
| 0.2748 | 2.6197 | 4500 | 0.3272 | 0.7660 | 0.7924 | 0.7790 | 0.9782 | |
|
| 0.2644 | 2.9108 | 5000 | 0.3192 | 0.7817 | 0.7811 | 0.7814 | 0.9779 | |
|
| 0.2416 | 3.2019 | 5500 | 0.3239 | 0.8004 | 0.7681 | 0.7839 | 0.9782 | |
|
| 0.2303 | 3.4929 | 6000 | 0.3085 | 0.7846 | 0.7966 | 0.7905 | 0.9787 | |
|
| 0.2252 | 3.7840 | 6500 | 0.3051 | 0.7973 | 0.7883 | 0.7928 | 0.9787 | |
|
| 0.2159 | 4.0751 | 7000 | 0.3045 | 0.7987 | 0.7908 | 0.7948 | 0.9790 | |
|
| 0.2067 | 4.3662 | 7500 | 0.2979 | 0.7969 | 0.7943 | 0.7956 | 0.9793 | |
|
| 0.2028 | 4.6573 | 8000 | 0.2924 | 0.7855 | 0.8132 | 0.7991 | 0.9792 | |
|
| 0.1985 | 4.9483 | 8500 | 0.2904 | 0.8008 | 0.7986 | 0.7997 | 0.9791 | |
|
| 0.1867 | 5.2394 | 9000 | 0.2884 | 0.8 | 0.8033 | 0.8017 | 0.9797 | |
|
| 0.1838 | 5.5305 | 9500 | 0.2841 | 0.7997 | 0.8220 | 0.8107 | 0.9800 | |
|
| 0.1838 | 5.8216 | 10000 | 0.2810 | 0.7895 | 0.8165 | 0.8028 | 0.9798 | |
|
| 0.1786 | 6.1126 | 10500 | 0.2767 | 0.8065 | 0.8150 | 0.8108 | 0.9802 | |
|
| 0.1719 | 6.4037 | 11000 | 0.2790 | 0.8133 | 0.8057 | 0.8095 | 0.9803 | |
|
| 0.1706 | 6.6948 | 11500 | 0.2795 | 0.8140 | 0.7983 | 0.8061 | 0.9802 | |
|
| 0.1695 | 6.9859 | 12000 | 0.2723 | 0.8124 | 0.8121 | 0.8123 | 0.9807 | |
|
| 0.1638 | 7.2770 | 12500 | 0.2726 | 0.8070 | 0.8078 | 0.8074 | 0.9803 | |
|
| 0.162 | 7.5680 | 13000 | 0.2724 | 0.8118 | 0.8173 | 0.8146 | 0.9807 | |
|
| 0.1619 | 7.8591 | 13500 | 0.2678 | 0.8018 | 0.8235 | 0.8125 | 0.9805 | |
|
| 0.1594 | 8.1502 | 14000 | 0.2719 | 0.8103 | 0.8068 | 0.8086 | 0.9800 | |
|
| 0.1571 | 8.4413 | 14500 | 0.2688 | 0.8097 | 0.8127 | 0.8112 | 0.9805 | |
|
| 0.1585 | 8.7324 | 15000 | 0.2673 | 0.8126 | 0.8150 | 0.8138 | 0.9806 | |
|
| 0.1546 | 9.0234 | 15500 | 0.2658 | 0.8105 | 0.8120 | 0.8112 | 0.9805 | |
|
| 0.1534 | 9.3145 | 16000 | 0.2652 | 0.8101 | 0.8198 | 0.8149 | 0.9807 | |
|
| 0.1535 | 9.6056 | 16500 | 0.2646 | 0.8097 | 0.8140 | 0.8119 | 0.9807 | |
|
| 0.1531 | 9.8967 | 17000 | 0.2650 | 0.8107 | 0.8117 | 0.8112 | 0.9806 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.19.1 |
|
|
