update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- glue
|
7 |
+
metrics:
|
8 |
+
- spearmanr
|
9 |
+
model-index:
|
10 |
+
- name: mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_stsb
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Text Classification
|
14 |
+
type: text-classification
|
15 |
+
dataset:
|
16 |
+
name: glue
|
17 |
+
type: glue
|
18 |
+
config: stsb
|
19 |
+
split: validation
|
20 |
+
args: stsb
|
21 |
+
metrics:
|
22 |
+
- name: Spearmanr
|
23 |
+
type: spearmanr
|
24 |
+
value: 0.8624005783710303
|
25 |
+
---
|
26 |
+
|
27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
+
should probably proofread and complete it, then remove this comment. -->
|
29 |
+
|
30 |
+
# mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_stsb
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the glue dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.2985
|
35 |
+
- Pearson: 0.8647
|
36 |
+
- Spearmanr: 0.8624
|
37 |
+
- Combined Score: 0.8636
|
38 |
+
|
39 |
+
## Model description
|
40 |
+
|
41 |
+
More information needed
|
42 |
+
|
43 |
+
## Intended uses & limitations
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Training and evaluation data
|
48 |
+
|
49 |
+
More information needed
|
50 |
+
|
51 |
+
## Training procedure
|
52 |
+
|
53 |
+
### Training hyperparameters
|
54 |
+
|
55 |
+
The following hyperparameters were used during training:
|
56 |
+
- learning_rate: 5e-05
|
57 |
+
- train_batch_size: 128
|
58 |
+
- eval_batch_size: 128
|
59 |
+
- seed: 10
|
60 |
+
- distributed_type: multi-GPU
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- num_epochs: 50
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|
68 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:---------:|:--------------:|
|
69 |
+
| 1.1501 | 1.0 | 45 | 0.4726 | 0.7774 | 0.7922 | 0.7848 |
|
70 |
+
| 0.364 | 2.0 | 90 | 0.3480 | 0.8457 | 0.8455 | 0.8456 |
|
71 |
+
| 0.259 | 3.0 | 135 | 0.3156 | 0.8582 | 0.8590 | 0.8586 |
|
72 |
+
| 0.2054 | 4.0 | 180 | 0.4231 | 0.8551 | 0.8549 | 0.8550 |
|
73 |
+
| 0.1629 | 5.0 | 225 | 0.3245 | 0.8668 | 0.8654 | 0.8661 |
|
74 |
+
| 0.1263 | 6.0 | 270 | 0.3192 | 0.8649 | 0.8625 | 0.8637 |
|
75 |
+
| 0.1021 | 7.0 | 315 | 0.3337 | 0.8655 | 0.8629 | 0.8642 |
|
76 |
+
| 0.0841 | 8.0 | 360 | 0.3061 | 0.8601 | 0.8577 | 0.8589 |
|
77 |
+
| 0.0713 | 9.0 | 405 | 0.3600 | 0.8576 | 0.8555 | 0.8566 |
|
78 |
+
| 0.0587 | 10.0 | 450 | 0.3135 | 0.8620 | 0.8600 | 0.8610 |
|
79 |
+
| 0.0488 | 11.0 | 495 | 0.3006 | 0.8641 | 0.8620 | 0.8631 |
|
80 |
+
| 0.0441 | 12.0 | 540 | 0.3308 | 0.8645 | 0.8621 | 0.8633 |
|
81 |
+
| 0.0385 | 13.0 | 585 | 0.3468 | 0.8620 | 0.8601 | 0.8610 |
|
82 |
+
| 0.0346 | 14.0 | 630 | 0.3175 | 0.8658 | 0.8634 | 0.8646 |
|
83 |
+
| 0.0298 | 15.0 | 675 | 0.2919 | 0.8665 | 0.8642 | 0.8654 |
|
84 |
+
| 0.0299 | 16.0 | 720 | 0.3103 | 0.8649 | 0.8628 | 0.8639 |
|
85 |
+
| 0.0263 | 17.0 | 765 | 0.3325 | 0.8620 | 0.8599 | 0.8609 |
|
86 |
+
| 0.0237 | 18.0 | 810 | 0.3092 | 0.8636 | 0.8611 | 0.8623 |
|
87 |
+
| 0.0213 | 19.0 | 855 | 0.3169 | 0.8653 | 0.8631 | 0.8642 |
|
88 |
+
| 0.0196 | 20.0 | 900 | 0.2985 | 0.8647 | 0.8624 | 0.8636 |
|
89 |
+
|
90 |
+
|
91 |
+
### Framework versions
|
92 |
+
|
93 |
+
- Transformers 4.26.0
|
94 |
+
- Pytorch 1.14.0a0+410ce96
|
95 |
+
- Datasets 2.9.0
|
96 |
+
- Tokenizers 0.13.2
|