update model card README.md
Browse files
README.md
CHANGED
|
@@ -1,8 +1,6 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
-
datasets:
|
| 5 |
-
- korquad
|
| 6 |
model-index:
|
| 7 |
- name: neg_komrc_train
|
| 8 |
results: []
|
|
@@ -13,9 +11,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# neg_komrc_train
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [beomi/kcbert-base](https://huggingface.co/beomi/kcbert-base) on the
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -47,49 +45,16 @@ The following hyperparameters were used during training:
|
|
| 47 |
|
| 48 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
|:-------------:|:-----:|:-----:|:---------------:|
|
| 50 |
-
|
|
| 51 |
-
|
|
| 52 |
-
|
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.8159 | 0.47 | 3000 | 0.7337 |
|
| 56 |
-
| 0.8064 | 0.55 | 3500 | 0.7185 |
|
| 57 |
-
| 0.765 | 0.63 | 4000 | 0.7556 |
|
| 58 |
-
| 0.74 | 0.7 | 4500 | 0.6704 |
|
| 59 |
-
| 0.7404 | 0.78 | 5000 | 0.7044 |
|
| 60 |
-
| 0.7118 | 0.86 | 5500 | 0.6737 |
|
| 61 |
-
| 0.6812 | 0.94 | 6000 | 0.6615 |
|
| 62 |
-
| 0.6193 | 1.02 | 6500 | 0.7373 |
|
| 63 |
-
| 0.4354 | 1.1 | 7000 | 0.6854 |
|
| 64 |
-
| 0.4415 | 1.17 | 7500 | 0.6764 |
|
| 65 |
-
| 0.4191 | 1.25 | 8000 | 0.6995 |
|
| 66 |
-
| 0.4292 | 1.33 | 8500 | 0.6724 |
|
| 67 |
-
| 0.4373 | 1.41 | 9000 | 0.7090 |
|
| 68 |
-
| 0.4082 | 1.49 | 9500 | 0.7398 |
|
| 69 |
-
| 0.4311 | 1.56 | 10000 | 0.6939 |
|
| 70 |
-
| 0.4068 | 1.64 | 10500 | 0.6926 |
|
| 71 |
-
| 0.4218 | 1.72 | 11000 | 0.6547 |
|
| 72 |
-
| 0.3853 | 1.8 | 11500 | 0.6734 |
|
| 73 |
-
| 0.4195 | 1.88 | 12000 | 0.6589 |
|
| 74 |
-
| 0.385 | 1.96 | 12500 | 0.6627 |
|
| 75 |
-
| 0.3267 | 2.03 | 13000 | 0.7509 |
|
| 76 |
-
| 0.2497 | 2.11 | 13500 | 0.7515 |
|
| 77 |
-
| 0.2406 | 2.19 | 14000 | 0.7753 |
|
| 78 |
-
| 0.246 | 2.27 | 14500 | 0.7827 |
|
| 79 |
-
| 0.2384 | 2.35 | 15000 | 0.7764 |
|
| 80 |
-
| 0.2274 | 2.42 | 15500 | 0.8102 |
|
| 81 |
-
| 0.2436 | 2.5 | 16000 | 0.7753 |
|
| 82 |
-
| 0.2275 | 2.58 | 16500 | 0.8086 |
|
| 83 |
-
| 0.2192 | 2.66 | 17000 | 0.8067 |
|
| 84 |
-
| 0.226 | 2.74 | 17500 | 0.7944 |
|
| 85 |
-
| 0.2305 | 2.82 | 18000 | 0.7867 |
|
| 86 |
-
| 0.2357 | 2.89 | 18500 | 0.7902 |
|
| 87 |
-
| 0.2343 | 2.97 | 19000 | 0.7822 |
|
| 88 |
|
| 89 |
|
| 90 |
### Framework versions
|
| 91 |
|
| 92 |
- Transformers 4.13.0
|
| 93 |
- Pytorch 1.10.0+cu111
|
| 94 |
-
- Datasets 1.18.
|
| 95 |
- Tokenizers 0.10.3
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
|
|
|
|
|
|
| 4 |
model-index:
|
| 5 |
- name: neg_komrc_train
|
| 6 |
results: []
|
|
|
|
| 11 |
|
| 12 |
# neg_komrc_train
|
| 13 |
|
| 14 |
+
This model is a fine-tuned version of [beomi/kcbert-base](https://huggingface.co/beomi/kcbert-base) on the None dataset.
|
| 15 |
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.4016
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 45 |
|
| 46 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 47 |
|:-------------:|:-----:|:-----:|:---------------:|
|
| 48 |
+
| 0.277 | 0.51 | 10000 | 0.4016 |
|
| 49 |
+
| 0.1671 | 1.03 | 20000 | 0.4116 |
|
| 50 |
+
| 0.1725 | 1.54 | 30000 | 0.4390 |
|
| 51 |
+
| 0.0868 | 2.06 | 40000 | 0.5147 |
|
| 52 |
+
| 0.0868 | 2.57 | 50000 | 0.5064 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
|
| 54 |
|
| 55 |
### Framework versions
|
| 56 |
|
| 57 |
- Transformers 4.13.0
|
| 58 |
- Pytorch 1.10.0+cu111
|
| 59 |
+
- Datasets 1.18.4
|
| 60 |
- Tokenizers 0.10.3
|