Commit
•
2a65515
1
Parent(s):
06f4770
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- precision
|
| 6 |
+
- recall
|
| 7 |
+
- f1
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: tabert-2k-naamapadam
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# tabert-2k-naamapadam
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [livinNector/tabert-2k](https://huggingface.co/livinNector/tabert-2k) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2850
|
| 22 |
+
- Precision: 0.7765
|
| 23 |
+
- Recall: 0.8041
|
| 24 |
+
- F1: 0.7901
|
| 25 |
+
- Accuracy: 0.9065
|
| 26 |
+
|
| 27 |
+
## Model description
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Intended uses & limitations
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training and evaluation data
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training procedure
|
| 40 |
+
|
| 41 |
+
### Training hyperparameters
|
| 42 |
+
|
| 43 |
+
The following hyperparameters were used during training:
|
| 44 |
+
- learning_rate: 5e-05
|
| 45 |
+
- train_batch_size: 64
|
| 46 |
+
- eval_batch_size: 128
|
| 47 |
+
- seed: 42
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: linear
|
| 50 |
+
- num_epochs: 2
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 55 |
+
|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 56 |
+
| 0.4679 | 0.05 | 400 | 0.3991 | 0.7155 | 0.6561 | 0.6845 | 0.8720 |
|
| 57 |
+
| 0.3907 | 0.1 | 800 | 0.3632 | 0.7181 | 0.7233 | 0.7207 | 0.8822 |
|
| 58 |
+
| 0.3663 | 0.15 | 1200 | 0.3483 | 0.7271 | 0.7371 | 0.7321 | 0.8857 |
|
| 59 |
+
| 0.3557 | 0.21 | 1600 | 0.3457 | 0.7286 | 0.7506 | 0.7395 | 0.8874 |
|
| 60 |
+
| 0.3533 | 0.26 | 2000 | 0.3413 | 0.7371 | 0.7435 | 0.7403 | 0.8895 |
|
| 61 |
+
| 0.3396 | 0.31 | 2400 | 0.3326 | 0.7435 | 0.7546 | 0.7490 | 0.8910 |
|
| 62 |
+
| 0.3302 | 0.36 | 2800 | 0.3264 | 0.7528 | 0.7553 | 0.7540 | 0.8937 |
|
| 63 |
+
| 0.3344 | 0.41 | 3200 | 0.3231 | 0.7503 | 0.7720 | 0.7610 | 0.8951 |
|
| 64 |
+
| 0.3262 | 0.46 | 3600 | 0.3228 | 0.7387 | 0.7762 | 0.7570 | 0.8941 |
|
| 65 |
+
| 0.3186 | 0.51 | 4000 | 0.3158 | 0.7699 | 0.7666 | 0.7683 | 0.8986 |
|
| 66 |
+
| 0.3163 | 0.57 | 4400 | 0.3130 | 0.7453 | 0.7798 | 0.7622 | 0.8955 |
|
| 67 |
+
| 0.3143 | 0.62 | 4800 | 0.3150 | 0.7572 | 0.7751 | 0.7660 | 0.8961 |
|
| 68 |
+
| 0.3088 | 0.67 | 5200 | 0.3151 | 0.7543 | 0.7828 | 0.7683 | 0.8972 |
|
| 69 |
+
| 0.3115 | 0.72 | 5600 | 0.3141 | 0.7708 | 0.7706 | 0.7707 | 0.8977 |
|
| 70 |
+
| 0.3095 | 0.77 | 6000 | 0.3043 | 0.7657 | 0.7831 | 0.7743 | 0.8991 |
|
| 71 |
+
| 0.3044 | 0.82 | 6400 | 0.3087 | 0.7526 | 0.7881 | 0.7699 | 0.8972 |
|
| 72 |
+
| 0.2964 | 0.87 | 6800 | 0.3070 | 0.7644 | 0.7928 | 0.7783 | 0.8992 |
|
| 73 |
+
| 0.2972 | 0.93 | 7200 | 0.3102 | 0.7692 | 0.7738 | 0.7715 | 0.8999 |
|
| 74 |
+
| 0.2985 | 0.98 | 7600 | 0.3016 | 0.7731 | 0.7858 | 0.7794 | 0.9018 |
|
| 75 |
+
| 0.2822 | 1.03 | 8000 | 0.3049 | 0.7734 | 0.7909 | 0.7820 | 0.9031 |
|
| 76 |
+
| 0.2764 | 1.08 | 8400 | 0.3059 | 0.7575 | 0.7976 | 0.7770 | 0.9011 |
|
| 77 |
+
| 0.2752 | 1.13 | 8800 | 0.3052 | 0.7553 | 0.7996 | 0.7768 | 0.9015 |
|
| 78 |
+
| 0.2689 | 1.18 | 9200 | 0.2990 | 0.7642 | 0.7982 | 0.7808 | 0.9037 |
|
| 79 |
+
| 0.2738 | 1.23 | 9600 | 0.2985 | 0.7698 | 0.7987 | 0.7840 | 0.9035 |
|
| 80 |
+
| 0.2731 | 1.29 | 10000 | 0.2950 | 0.7713 | 0.7982 | 0.7845 | 0.9037 |
|
| 81 |
+
| 0.2694 | 1.34 | 10400 | 0.2920 | 0.7743 | 0.8017 | 0.7878 | 0.9059 |
|
| 82 |
+
| 0.2727 | 1.39 | 10800 | 0.2931 | 0.7693 | 0.7979 | 0.7834 | 0.9040 |
|
| 83 |
+
| 0.2622 | 1.44 | 11200 | 0.2946 | 0.7702 | 0.7942 | 0.7820 | 0.9032 |
|
| 84 |
+
| 0.2672 | 1.49 | 11600 | 0.2894 | 0.7724 | 0.8062 | 0.7890 | 0.9060 |
|
| 85 |
+
| 0.2601 | 1.54 | 12000 | 0.2907 | 0.7706 | 0.8010 | 0.7855 | 0.9058 |
|
| 86 |
+
| 0.2629 | 1.59 | 12400 | 0.2930 | 0.7628 | 0.8150 | 0.7880 | 0.9052 |
|
| 87 |
+
| 0.2635 | 1.65 | 12800 | 0.2907 | 0.7775 | 0.7970 | 0.7871 | 0.9047 |
|
| 88 |
+
| 0.2673 | 1.7 | 13200 | 0.2909 | 0.7753 | 0.7982 | 0.7866 | 0.9045 |
|
| 89 |
+
| 0.2726 | 1.75 | 13600 | 0.2880 | 0.7714 | 0.8048 | 0.7877 | 0.9054 |
|
| 90 |
+
| 0.2607 | 1.8 | 14000 | 0.2850 | 0.7760 | 0.8010 | 0.7883 | 0.9053 |
|
| 91 |
+
| 0.2684 | 1.85 | 14400 | 0.2847 | 0.7709 | 0.8077 | 0.7889 | 0.9059 |
|
| 92 |
+
| 0.2625 | 1.9 | 14800 | 0.2849 | 0.7742 | 0.8079 | 0.7907 | 0.9067 |
|
| 93 |
+
| 0.2631 | 1.95 | 15200 | 0.2850 | 0.7765 | 0.8041 | 0.7901 | 0.9065 |
|
| 94 |
+
|
| 95 |
+
|
| 96 |
+
### Framework versions
|
| 97 |
+
|
| 98 |
+
- Transformers 4.29.2
|
| 99 |
+
- Pytorch 2.0.0
|
| 100 |
+
- Datasets 2.12.0
|
| 101 |
+
- Tokenizers 0.13.3
|