asadfgglie
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -1,70 +1,87 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
base_model: MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
4 |
-
tags:
|
5 |
-
- generated_from_trainer
|
6 |
-
metrics:
|
7 |
-
- accuracy
|
8 |
-
model-index:
|
9 |
-
- name: mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
10 |
-
results: []
|
11 |
-
---
|
12 |
-
|
13 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
-
should probably proofread and complete it, then remove this comment. -->
|
15 |
-
|
16 |
-
# mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
17 |
-
|
18 |
-
This model is a fine-tuned version of [MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7](https://huggingface.co/MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7) on the None dataset.
|
19 |
-
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.2718
|
21 |
-
- F1 Macro: 0.9088
|
22 |
-
- F1 Micro: 0.9089
|
23 |
-
- Accuracy Balanced: 0.9089
|
24 |
-
- Accuracy: 0.9089
|
25 |
-
- Precision Macro: 0.9092
|
26 |
-
- Recall Macro: 0.9089
|
27 |
-
- Precision Micro: 0.9089
|
28 |
-
- Recall Micro: 0.9089
|
29 |
-
|
30 |
-
## Model description
|
31 |
-
|
32 |
-
More information needed
|
33 |
-
|
34 |
-
## Intended uses & limitations
|
35 |
-
|
36 |
-
More information needed
|
37 |
-
|
38 |
-
## Training and evaluation data
|
39 |
-
|
40 |
-
More information needed
|
41 |
-
|
42 |
-
## Training procedure
|
43 |
-
|
44 |
-
### Training hyperparameters
|
45 |
-
|
46 |
-
The following hyperparameters were used during training:
|
47 |
-
- learning_rate: 2e-05
|
48 |
-
- train_batch_size: 16
|
49 |
-
- eval_batch_size: 128
|
50 |
-
- seed: 20241201
|
51 |
-
- gradient_accumulation_steps: 2
|
52 |
-
- total_train_batch_size: 32
|
53 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
54 |
-
- lr_scheduler_type: linear
|
55 |
-
- lr_scheduler_warmup_ratio: 0.06
|
56 |
-
- num_epochs: 3
|
57 |
-
|
58 |
-
### Training results
|
59 |
-
|
60 |
-
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Micro | Accuracy Balanced | Accuracy | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
|
61 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------------:|:--------:|:---------------:|:------------:|:---------------:|:------------:|
|
62 |
-
| 0.2798 | 1.69 | 200 | 0.3328 | 0.8677 | 0.8677 | 0.8681 | 0.8677 | 0.8678 | 0.8681 | 0.8677 | 0.8677 |
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
base_model: MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7](https://huggingface.co/MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7) on the None dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.2718
|
21 |
+
- F1 Macro: 0.9088
|
22 |
+
- F1 Micro: 0.9089
|
23 |
+
- Accuracy Balanced: 0.9089
|
24 |
+
- Accuracy: 0.9089
|
25 |
+
- Precision Macro: 0.9092
|
26 |
+
- Recall Macro: 0.9089
|
27 |
+
- Precision Micro: 0.9089
|
28 |
+
- Recall Micro: 0.9089
|
29 |
+
|
30 |
+
## Model description
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Intended uses & limitations
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training and evaluation data
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Training procedure
|
43 |
+
|
44 |
+
### Training hyperparameters
|
45 |
+
|
46 |
+
The following hyperparameters were used during training:
|
47 |
+
- learning_rate: 2e-05
|
48 |
+
- train_batch_size: 16
|
49 |
+
- eval_batch_size: 128
|
50 |
+
- seed: 20241201
|
51 |
+
- gradient_accumulation_steps: 2
|
52 |
+
- total_train_batch_size: 32
|
53 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
54 |
+
- lr_scheduler_type: linear
|
55 |
+
- lr_scheduler_warmup_ratio: 0.06
|
56 |
+
- num_epochs: 3
|
57 |
+
|
58 |
+
### Training results
|
59 |
+
|
60 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Micro | Accuracy Balanced | Accuracy | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
|
61 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------------:|:--------:|:---------------:|:------------:|:---------------:|:------------:|
|
62 |
+
| 0.2798 | 1.69 | 200 | 0.3328 | 0.8677 | 0.8677 | 0.8681 | 0.8677 | 0.8678 | 0.8681 | 0.8677 | 0.8677 |
|
63 |
+
|
64 |
+
### Eval results
|
65 |
+
|Datasets|asadfgglie/nli-zh-tw-all/test|asadfgglie/BanBan_2024-10-17-facial_expressions-nli/test|eval_dataset|test_dataset|
|
66 |
+
| :---: | :---: | :---: | :---: | :---: |
|
67 |
+
|eval_loss|0.667|0.294|0.381|0.272|
|
68 |
+
|eval_f1_macro|0.711|0.901|0.868|0.909|
|
69 |
+
|eval_f1_micro|0.713|0.901|0.868|0.909|
|
70 |
+
|eval_accuracy_balanced|0.71|0.901|0.867|0.909|
|
71 |
+
|eval_accuracy|0.713|0.901|0.868|0.909|
|
72 |
+
|eval_precision_macro|0.711|0.901|0.868|0.909|
|
73 |
+
|eval_recall_macro|0.71|0.901|0.867|0.909|
|
74 |
+
|eval_precision_micro|0.713|0.901|0.868|0.909|
|
75 |
+
|eval_recall_micro|0.713|0.901|0.868|0.909|
|
76 |
+
|eval_runtime|568.387|4.571|0.829|3.382|
|
77 |
+
|eval_samples_per_second|14.955|206.945|227.909|223.805|
|
78 |
+
|eval_steps_per_second|0.118|1.75|2.412|1.774|
|
79 |
+
|epoch|2.99|2.99|2.99|2.99|
|
80 |
+
|Size of dataset|8500|946|189|757|
|
81 |
+
|
82 |
+
### Framework versions
|
83 |
+
|
84 |
+
- Transformers 4.33.3
|
85 |
+
- Pytorch 2.5.1+cu121
|
86 |
+
- Datasets 2.14.7
|
87 |
+
- Tokenizers 0.13.3
|