File size: 3,625 Bytes
22a727b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 |
---
base_model: allenai/scibert_scivocab_uncased
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: SciBERT_100K_steps
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. -->
# SciBERT_100K_steps
This model is a fine-tuned version of [allenai/scibert_scivocab_uncased](https://huggingface.co/allenai/scibert_scivocab_uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0144
- Accuracy: 0.9947
- Precision: 0.7850
- Recall: 0.6355
- F1: 0.7024
- Hamming: 0.0053
## 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
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 100000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Hamming |
|:-------------:|:-----:|:------:|:---------------:|:--------:|:---------:|:------:|:------:|:-------:|
| 0.1681 | 0.08 | 5000 | 0.0487 | 0.9902 | 0.0 | 0.0 | 0.0 | 0.0098 |
| 0.032 | 0.16 | 10000 | 0.0223 | 0.9930 | 0.8068 | 0.3728 | 0.5100 | 0.0070 |
| 0.0201 | 0.24 | 15000 | 0.0186 | 0.9937 | 0.7815 | 0.4970 | 0.6076 | 0.0063 |
| 0.018 | 0.32 | 20000 | 0.0172 | 0.9941 | 0.7763 | 0.5550 | 0.6472 | 0.0059 |
| 0.017 | 0.4 | 25000 | 0.0166 | 0.9942 | 0.7864 | 0.5624 | 0.6558 | 0.0058 |
| 0.0166 | 0.47 | 30000 | 0.0163 | 0.9943 | 0.7707 | 0.5880 | 0.6671 | 0.0057 |
| 0.0163 | 0.55 | 35000 | 0.0160 | 0.9943 | 0.7802 | 0.5809 | 0.6659 | 0.0057 |
| 0.0159 | 0.63 | 40000 | 0.0158 | 0.9944 | 0.7719 | 0.6012 | 0.6759 | 0.0056 |
| 0.0157 | 0.71 | 45000 | 0.0155 | 0.9945 | 0.7750 | 0.6104 | 0.6829 | 0.0055 |
| 0.0154 | 0.79 | 50000 | 0.0153 | 0.9945 | 0.7734 | 0.6202 | 0.6884 | 0.0055 |
| 0.0153 | 0.87 | 55000 | 0.0151 | 0.9945 | 0.7823 | 0.6072 | 0.6837 | 0.0055 |
| 0.0152 | 0.95 | 60000 | 0.0151 | 0.9945 | 0.7813 | 0.6124 | 0.6866 | 0.0055 |
| 0.0148 | 1.03 | 65000 | 0.0149 | 0.9946 | 0.7843 | 0.6208 | 0.6930 | 0.0054 |
| 0.0143 | 1.11 | 70000 | 0.0148 | 0.9946 | 0.7802 | 0.6231 | 0.6929 | 0.0054 |
| 0.0142 | 1.19 | 75000 | 0.0148 | 0.9946 | 0.7714 | 0.6377 | 0.6982 | 0.0054 |
| 0.0141 | 1.27 | 80000 | 0.0146 | 0.9947 | 0.7837 | 0.6281 | 0.6973 | 0.0053 |
| 0.0141 | 1.34 | 85000 | 0.0146 | 0.9947 | 0.7836 | 0.6374 | 0.7030 | 0.0053 |
| 0.014 | 1.42 | 90000 | 0.0145 | 0.9947 | 0.7859 | 0.6326 | 0.7010 | 0.0053 |
| 0.0139 | 1.5 | 95000 | 0.0145 | 0.9947 | 0.7875 | 0.6317 | 0.7010 | 0.0053 |
| 0.0139 | 1.58 | 100000 | 0.0144 | 0.9947 | 0.7850 | 0.6355 | 0.7024 | 0.0053 |
### Framework versions
- Transformers 4.35.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.7.1
- Tokenizers 0.14.1
|