Hugging Face's logo

haryoaw
/
scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_

Transformers
PyTorch
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Community
haryoaw commited on
Commit
1e85a2d
•
1 Parent(s): 11cacf6

Initial Commit

Browse files
Files changed (4) hide show
  1. README.md +51 -67
  2. config.json +3 -3
  3. pytorch_model.bin +2 -2
  4. training_args.bin +1 -1
README.md CHANGED
@@ -10,23 +10,7 @@ metrics:
10
  - f1
11
  model-index:
12
  - name: scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_
13
- results:
14
- - task:
15
- name: Text Classification
16
- type: text-classification
17
- dataset:
18
- name: tweet_sentiment_multilingual
19
- type: tweet_sentiment_multilingual
20
- config: all
21
- split: validation
22
- args: all
23
- metrics:
24
- - name: Accuracy
25
- type: accuracy
26
- value: 0.589891975308642
27
- - name: F1
28
- type: f1
29
- value: 0.588413122388427
30
  ---
31
 
32
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -36,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
36
 
37
  This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the tweet_sentiment_multilingual dataset.
38
  It achieves the following results on the evaluation set:
39
- - Loss: 4.1967
40
- - Accuracy: 0.5899
41
- - F1: 0.5884
42
 
43
  ## Model description
44
 
@@ -60,7 +44,7 @@ The following hyperparameters were used during training:
60
  - learning_rate: 5e-05
61
  - train_batch_size: 32
62
  - eval_batch_size: 32
63
- - seed: 11213
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
  - num_epochs: 50
@@ -69,52 +53,52 @@ The following hyperparameters were used during training:
69
 
70
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
71
  |:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|
72
- | 1.0743 | 1.09 | 500 | 1.0031 | 0.5066 | 0.5042 |
73
- | 0.9346 | 2.17 | 1000 | 0.9437 | 0.5656 | 0.5679 |
74
- | 0.8123 | 3.26 | 1500 | 0.9108 | 0.5949 | 0.5905 |
75
- | 0.6842 | 4.35 | 2000 | 1.1082 | 0.5756 | 0.5661 |
76
- | 0.5603 | 5.43 | 2500 | 1.1812 | 0.5907 | 0.5828 |
77
- | 0.4284 | 6.52 | 3000 | 1.3230 | 0.5895 | 0.5870 |
78
- | 0.3295 | 7.61 | 3500 | 1.4855 | 0.5637 | 0.5638 |
79
- | 0.2589 | 8.7 | 4000 | 1.5869 | 0.5837 | 0.5784 |
80
- | 0.2035 | 9.78 | 4500 | 1.8098 | 0.5826 | 0.5776 |
81
- | 0.1755 | 10.87 | 5000 | 1.7393 | 0.5887 | 0.5856 |
82
- | 0.1497 | 11.96 | 5500 | 2.1213 | 0.5887 | 0.5828 |
83
- | 0.13 | 13.04 | 6000 | 2.2126 | 0.5833 | 0.5827 |
84
- | 0.1151 | 14.13 | 6500 | 2.2685 | 0.5818 | 0.5811 |
85
- | 0.1028 | 15.22 | 7000 | 2.5633 | 0.5826 | 0.5827 |
86
- | 0.0962 | 16.3 | 7500 | 2.4350 | 0.5795 | 0.5770 |
87
- | 0.0804 | 17.39 | 8000 | 2.6830 | 0.5806 | 0.5752 |
88
- | 0.0781 | 18.48 | 8500 | 2.6389 | 0.5818 | 0.5811 |
89
- | 0.0677 | 19.57 | 9000 | 2.6490 | 0.5806 | 0.5788 |
90
- | 0.0593 | 20.65 | 9500 | 2.9908 | 0.5768 | 0.5732 |
91
- | 0.0578 | 21.74 | 10000 | 2.9127 | 0.5845 | 0.5828 |
92
- | 0.0493 | 22.83 | 10500 | 3.0101 | 0.5802 | 0.5744 |
93
- | 0.0455 | 23.91 | 11000 | 2.9419 | 0.5795 | 0.5779 |
94
- | 0.0351 | 25.0 | 11500 | 3.2339 | 0.5752 | 0.5742 |
95
- | 0.0369 | 26.09 | 12000 | 3.2997 | 0.5899 | 0.5818 |
96
- | 0.0291 | 27.17 | 12500 | 3.5819 | 0.5833 | 0.5804 |
97
- | 0.0281 | 28.26 | 13000 | 3.4498 | 0.5795 | 0.5798 |
98
- | 0.0258 | 29.35 | 13500 | 3.5006 | 0.5768 | 0.5768 |
99
- | 0.027 | 30.43 | 14000 | 3.4740 | 0.5849 | 0.5832 |
100
- | 0.0218 | 31.52 | 14500 | 3.2293 | 0.5918 | 0.5907 |
101
- | 0.0227 | 32.61 | 15000 | 3.4840 | 0.5876 | 0.5861 |
102
- | 0.0212 | 33.7 | 15500 | 3.2922 | 0.5845 | 0.5841 |
103
- | 0.0119 | 34.78 | 16000 | 3.9035 | 0.5729 | 0.5744 |
104
- | 0.019 | 35.87 | 16500 | 3.5470 | 0.5795 | 0.5781 |
105
- | 0.0146 | 36.96 | 17000 | 3.7651 | 0.5795 | 0.5772 |
106
- | 0.0144 | 38.04 | 17500 | 3.7248 | 0.5829 | 0.5787 |
107
- | 0.0077 | 39.13 | 18000 | 4.1509 | 0.5806 | 0.5754 |
108
- | 0.0097 | 40.22 | 18500 | 3.8829 | 0.5829 | 0.5796 |
109
- | 0.0092 | 41.3 | 19000 | 3.8987 | 0.5853 | 0.5842 |
110
- | 0.0087 | 42.39 | 19500 | 3.8544 | 0.5899 | 0.5882 |
111
- | 0.0083 | 43.48 | 20000 | 3.9211 | 0.5895 | 0.5855 |
112
- | 0.006 | 44.57 | 20500 | 3.9856 | 0.5868 | 0.5856 |
113
- | 0.0062 | 45.65 | 21000 | 4.0873 | 0.5891 | 0.5872 |
114
- | 0.0027 | 46.74 | 21500 | 4.1639 | 0.5891 | 0.5888 |
115
- | 0.0052 | 47.83 | 22000 | 4.1754 | 0.5914 | 0.5893 |
116
- | 0.0031 | 48.91 | 22500 | 4.1914 | 0.5887 | 0.5879 |
117
- | 0.0035 | 50.0 | 23000 | 4.1967 | 0.5899 | 0.5884 |
118
 
119
 
120
  ### Framework versions
 
10
  - f1
11
  model-index:
12
  - name: scenario-NON-KD-PR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_
13
+ results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
20
 
21
  This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the tweet_sentiment_multilingual dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 3.2776
24
+ - Accuracy: 0.5490
25
+ - F1: 0.5470
26
 
27
  ## Model description
28
 
 
44
  - learning_rate: 5e-05
45
  - train_batch_size: 32
46
  - eval_batch_size: 32
47
+ - seed: 333
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
  - num_epochs: 50
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
55
  |:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|
56
+ | 1.0901 | 1.09 | 500 | 1.0564 | 0.4379 | 0.4046 |
57
+ | 1.0001 | 2.17 | 1000 | 1.0287 | 0.5085 | 0.4941 |
58
+ | 0.9108 | 3.26 | 1500 | 1.0254 | 0.5316 | 0.5273 |
59
+ | 0.8453 | 4.35 | 2000 | 0.9739 | 0.5390 | 0.5363 |
60
+ | 0.786 | 5.43 | 2500 | 0.9965 | 0.5540 | 0.5500 |
61
+ | 0.7317 | 6.52 | 3000 | 1.0309 | 0.5505 | 0.5452 |
62
+ | 0.6714 | 7.61 | 3500 | 1.1479 | 0.5444 | 0.5466 |
63
+ | 0.6192 | 8.7 | 4000 | 1.0839 | 0.5536 | 0.5533 |
64
+ | 0.5693 | 9.78 | 4500 | 1.2411 | 0.5382 | 0.5259 |
65
+ | 0.5114 | 10.87 | 5000 | 1.2202 | 0.5486 | 0.5502 |
66
+ | 0.4705 | 11.96 | 5500 | 1.4185 | 0.5478 | 0.5445 |
67
+ | 0.425 | 13.04 | 6000 | 1.3994 | 0.5417 | 0.5314 |
68
+ | 0.3815 | 14.13 | 6500 | 1.5880 | 0.5475 | 0.5475 |
69
+ | 0.3405 | 15.22 | 7000 | 1.5789 | 0.5405 | 0.5330 |
70
+ | 0.3046 | 16.3 | 7500 | 1.7872 | 0.5405 | 0.5328 |
71
+ | 0.279 | 17.39 | 8000 | 1.7094 | 0.5417 | 0.5390 |
72
+ | 0.2488 | 18.48 | 8500 | 1.7790 | 0.5471 | 0.5451 |
73
+ | 0.2203 | 19.57 | 9000 | 1.8204 | 0.5478 | 0.5464 |
74
+ | 0.2145 | 20.65 | 9500 | 1.9339 | 0.5448 | 0.5386 |
75
+ | 0.1869 | 21.74 | 10000 | 2.1092 | 0.5390 | 0.5360 |
76
+ | 0.1788 | 22.83 | 10500 | 1.9770 | 0.5540 | 0.5513 |
77
+ | 0.1473 | 23.91 | 11000 | 2.1967 | 0.5471 | 0.5425 |
78
+ | 0.1437 | 25.0 | 11500 | 2.1961 | 0.5513 | 0.5431 |
79
+ | 0.1296 | 26.09 | 12000 | 2.2828 | 0.5536 | 0.5518 |
80
+ | 0.1151 | 27.17 | 12500 | 2.3900 | 0.5405 | 0.5346 |
81
+ | 0.1151 | 28.26 | 13000 | 2.5206 | 0.5440 | 0.5394 |
82
+ | 0.1058 | 29.35 | 13500 | 2.5638 | 0.5463 | 0.5413 |
83
+ | 0.1056 | 30.43 | 14000 | 2.6504 | 0.5417 | 0.5351 |
84
+ | 0.098 | 31.52 | 14500 | 2.6291 | 0.5571 | 0.5544 |
85
+ | 0.0918 | 32.61 | 15000 | 2.6844 | 0.5421 | 0.5408 |
86
+ | 0.0873 | 33.7 | 15500 | 2.7813 | 0.5401 | 0.5403 |
87
+ | 0.0897 | 34.78 | 16000 | 2.8257 | 0.5459 | 0.5428 |
88
+ | 0.0781 | 35.87 | 16500 | 2.8813 | 0.5478 | 0.5450 |
89
+ | 0.0698 | 36.96 | 17000 | 3.0486 | 0.5336 | 0.5303 |
90
+ | 0.0674 | 38.04 | 17500 | 3.1261 | 0.5475 | 0.5417 |
91
+ | 0.0756 | 39.13 | 18000 | 3.0463 | 0.5482 | 0.5480 |
92
+ | 0.0592 | 40.22 | 18500 | 3.1190 | 0.5440 | 0.5412 |
93
+ | 0.0562 | 41.3 | 19000 | 3.1770 | 0.5370 | 0.5342 |
94
+ | 0.0575 | 42.39 | 19500 | 3.1928 | 0.5432 | 0.5405 |
95
+ | 0.0534 | 43.48 | 20000 | 3.2141 | 0.5494 | 0.5462 |
96
+ | 0.0487 | 44.57 | 20500 | 3.2784 | 0.5440 | 0.5376 |
97
+ | 0.0472 | 45.65 | 21000 | 3.2675 | 0.5451 | 0.5420 |
98
+ | 0.0495 | 46.74 | 21500 | 3.2487 | 0.5502 | 0.5474 |
99
+ | 0.0411 | 47.83 | 22000 | 3.2628 | 0.5486 | 0.5468 |
100
+ | 0.0417 | 48.91 | 22500 | 3.2780 | 0.5494 | 0.5476 |
101
+ | 0.0412 | 50.0 | 23000 | 3.2776 | 0.5490 | 0.5470 |
102
 
103
 
104
  ### Framework versions
config.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "_name_or_path": "xlm-roberta-base",
3
  "architectures": [
4
- "XLMRobertaForSequenceClassification"
5
  ],
6
  "attention_probs_dropout_prob": 0.1,
7
  "bos_token_id": 0,
@@ -9,14 +9,14 @@
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
  "hidden_dropout_prob": 0.1,
12
- "hidden_size": 768,
13
  "id2label": {
14
  "0": "LABEL_0",
15
  "1": "LABEL_1",
16
  "2": "LABEL_2"
17
  },
18
  "initializer_range": 0.02,
19
- "intermediate_size": 3072,
20
  "label2id": {
21
  "LABEL_0": 0,
22
  "LABEL_1": 1,
 
1
  {
2
  "_name_or_path": "xlm-roberta-base",
3
  "architectures": [
4
+ "XLMRobertaForSequenceClassificationKD"
5
  ],
6
  "attention_probs_dropout_prob": 0.1,
7
  "bos_token_id": 0,
 
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
  "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 384,
13
  "id2label": {
14
  "0": "LABEL_0",
15
  "1": "LABEL_1",
16
  "2": "LABEL_2"
17
  },
18
  "initializer_range": 0.02,
19
+ "intermediate_size": 1536,
20
  "label2id": {
21
  "LABEL_0": 0,
22
  "LABEL_1": 1,
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f2b00f463126da80528767f9f7ba7b18d771484abce738c5d2b1e8bbbee1f62a
3
- size 942111086
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f105078f6541c246d9a0ab2cd1e3b7423591a6a6fa7484509b8fa313d42f4d25
3
+ size 429199798
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40174e22a5a2e05c1689834ecc61f6377ae5b665a9c88a2c3d896fcd77e524bd
3
  size 4664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:166ec74f89602a4d853987af4b4f99ed1bccaff0844ae7d838470c6f12d39ac9
3
  size 4664