update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- accuracy
|
6 |
+
- f1
|
7 |
+
model-index:
|
8 |
+
- name: electra-large-discriminator-nli-efl-hateval
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# electra-large-discriminator-nli-efl-hateval
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [ynie/electra-large-discriminator-snli_mnli_fever_anli_R1_R2_R3-nli](https://huggingface.co/ynie/electra-large-discriminator-snli_mnli_fever_anli_R1_R2_R3-nli) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.4166
|
20 |
+
- Accuracy: 0.798
|
21 |
+
- F1: 0.7968
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 1e-06
|
41 |
+
- train_batch_size: 8
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- gradient_accumulation_steps: 16
|
45 |
+
- total_train_batch_size: 128
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- num_epochs: 20
|
49 |
+
|
50 |
+
### Training results
|
51 |
+
|
52 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|
53 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
|
54 |
+
| 0.4175 | 1.0 | 210 | 0.4020 | 0.7317 | 0.7305 |
|
55 |
+
| 0.3061 | 2.0 | 420 | 0.3520 | 0.768 | 0.7675 |
|
56 |
+
| 0.2588 | 3.0 | 630 | 0.3253 | 0.79 | 0.7888 |
|
57 |
+
| 0.234 | 4.0 | 840 | 0.3373 | 0.788 | 0.7877 |
|
58 |
+
| 0.2116 | 5.0 | 1050 | 0.3247 | 0.804 | 0.8033 |
|
59 |
+
| 0.1974 | 6.0 | 1260 | 0.3400 | 0.793 | 0.7928 |
|
60 |
+
| 0.1807 | 7.0 | 1470 | 0.3511 | 0.7973 | 0.7969 |
|
61 |
+
| 0.1715 | 8.0 | 1680 | 0.3496 | 0.7993 | 0.7989 |
|
62 |
+
| 0.1577 | 9.0 | 1890 | 0.3507 | 0.8043 | 0.8032 |
|
63 |
+
| 0.1469 | 10.0 | 2100 | 0.3604 | 0.798 | 0.7970 |
|
64 |
+
| 0.1394 | 11.0 | 2310 | 0.3734 | 0.7967 | 0.7957 |
|
65 |
+
| 0.1322 | 12.0 | 2520 | 0.3929 | 0.7913 | 0.7906 |
|
66 |
+
| 0.1231 | 13.0 | 2730 | 0.3954 | 0.795 | 0.7941 |
|
67 |
+
| 0.1189 | 14.0 | 2940 | 0.3994 | 0.7977 | 0.7963 |
|
68 |
+
| 0.1143 | 15.0 | 3150 | 0.3995 | 0.7993 | 0.7980 |
|
69 |
+
| 0.1083 | 16.0 | 3360 | 0.4125 | 0.7927 | 0.7918 |
|
70 |
+
| 0.1079 | 17.0 | 3570 | 0.4036 | 0.7993 | 0.7979 |
|
71 |
+
| 0.1055 | 18.0 | 3780 | 0.4121 | 0.7967 | 0.7956 |
|
72 |
+
| 0.1006 | 19.0 | 3990 | 0.4152 | 0.7973 | 0.7961 |
|
73 |
+
| 0.101 | 20.0 | 4200 | 0.4166 | 0.798 | 0.7968 |
|
74 |
+
|
75 |
+
|
76 |
+
### Framework versions
|
77 |
+
|
78 |
+
- Transformers 4.17.0
|
79 |
+
- Pytorch 1.11.0+cu113
|
80 |
+
- Datasets 2.0.0
|
81 |
+
- Tokenizers 0.11.6
|