update model card README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,4 @@
|
|
1 |
---
|
2 |
-
language:
|
3 |
-
- en
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
@@ -14,7 +12,7 @@ model-index:
|
|
14 |
name: Text Classification
|
15 |
type: text-classification
|
16 |
dataset:
|
17 |
-
name:
|
18 |
type: glue
|
19 |
config: qnli
|
20 |
split: validation
|
@@ -30,7 +28,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
30 |
|
31 |
# hBERTv1_no_pretrain_qnli
|
32 |
|
33 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on the
|
34 |
It achieves the following results on the evaluation set:
|
35 |
- Loss: 0.6931
|
36 |
- Accuracy: 0.5054
|
@@ -52,41 +50,37 @@ More information needed
|
|
52 |
### Training hyperparameters
|
53 |
|
54 |
The following hyperparameters were used during training:
|
55 |
-
- learning_rate:
|
56 |
-
- train_batch_size:
|
57 |
-
- eval_batch_size:
|
58 |
- seed: 10
|
59 |
- distributed_type: multi-GPU
|
60 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
- lr_scheduler_type: linear
|
62 |
- num_epochs: 50
|
63 |
-
- mixed_precision_training: Native AMP
|
64 |
|
65 |
### Training results
|
66 |
|
67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
68 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
-
| 0.
|
78 |
-
| 0.
|
79 |
-
| 0.6932 | 11.0 |
|
80 |
-
| 0.6932 | 12.0 |
|
81 |
-
| 0.
|
82 |
-
| 0.6932 | 14.0 | 11466 | 0.6931 | 0.4946 |
|
83 |
-
| 0.6932 | 15.0 | 12285 | 0.6934 | 0.4946 |
|
84 |
-
| 0.6932 | 16.0 | 13104 | 0.6931 | 0.4946 |
|
85 |
|
86 |
|
87 |
### Framework versions
|
88 |
|
89 |
-
- Transformers 4.
|
90 |
- Pytorch 1.14.0a0+410ce96
|
91 |
- Datasets 2.12.0
|
92 |
- Tokenizers 0.13.3
|
|
|
1 |
---
|
|
|
|
|
2 |
tags:
|
3 |
- generated_from_trainer
|
4 |
datasets:
|
|
|
12 |
name: Text Classification
|
13 |
type: text-classification
|
14 |
dataset:
|
15 |
+
name: glue
|
16 |
type: glue
|
17 |
config: qnli
|
18 |
split: validation
|
|
|
28 |
|
29 |
# hBERTv1_no_pretrain_qnli
|
30 |
|
31 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
|
32 |
It achieves the following results on the evaluation set:
|
33 |
- Loss: 0.6931
|
34 |
- Accuracy: 0.5054
|
|
|
50 |
### Training hyperparameters
|
51 |
|
52 |
The following hyperparameters were used during training:
|
53 |
+
- learning_rate: 4e-05
|
54 |
+
- train_batch_size: 96
|
55 |
+
- eval_batch_size: 96
|
56 |
- seed: 10
|
57 |
- distributed_type: multi-GPU
|
58 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
- lr_scheduler_type: linear
|
60 |
- num_epochs: 50
|
|
|
61 |
|
62 |
### Training results
|
63 |
|
64 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
65 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
66 |
+
| 0.7059 | 1.0 | 1092 | 0.7004 | 0.5054 |
|
67 |
+
| 0.6948 | 2.0 | 2184 | 0.6938 | 0.4946 |
|
68 |
+
| 0.6939 | 3.0 | 3276 | 0.6932 | 0.5054 |
|
69 |
+
| 0.6936 | 4.0 | 4368 | 0.6931 | 0.5054 |
|
70 |
+
| 0.6934 | 5.0 | 5460 | 0.6931 | 0.5054 |
|
71 |
+
| 0.6936 | 6.0 | 6552 | 0.6931 | 0.5054 |
|
72 |
+
| 0.6933 | 7.0 | 7644 | 0.6931 | 0.5054 |
|
73 |
+
| 0.6933 | 8.0 | 8736 | 0.6931 | 0.5054 |
|
74 |
+
| 0.6934 | 9.0 | 9828 | 0.6934 | 0.5054 |
|
75 |
+
| 0.6933 | 10.0 | 10920 | 0.6931 | 0.5054 |
|
76 |
+
| 0.6932 | 11.0 | 12012 | 0.6933 | 0.4946 |
|
77 |
+
| 0.6932 | 12.0 | 13104 | 0.6931 | 0.5054 |
|
78 |
+
| 0.6933 | 13.0 | 14196 | 0.6931 | 0.5054 |
|
|
|
|
|
|
|
79 |
|
80 |
|
81 |
### Framework versions
|
82 |
|
83 |
+
- Transformers 4.30.2
|
84 |
- Pytorch 1.14.0a0+410ce96
|
85 |
- Datasets 2.12.0
|
86 |
- Tokenizers 0.13.3
|