Hugging Face's logo

caush
/
Clickbait1

Text Classification
Transformers
PyTorch
TensorBoard
bert
Generated from Trainer
text-embeddings-inference
Model card Files
xet
Metrics Community
2
caush commited on
Commit
32b904d
·
1 Parent(s): 6dc2934

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +62 -55
README.md CHANGED
@@ -7,30 +7,28 @@ model-index:
7
  results: []
8
  ---
9
 
 
 
 
10
  # Clickbait1
11
 
12
- This model is a fine-tuned version of [microsoft/Multilingual-MiniLM-L12-H384](https://huggingface.co/microsoft/Multilingual-MiniLM-L12-H384) on the [Webis-Clickbait-17](https://zenodo.org/record/5530410) dataset.
13
  It achieves the following results on the evaluation set:
14
- - Loss: 0.0260
15
 
16
  ## Model description
17
 
18
- MiniLM is a distilled model from the paper "MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers".
19
-
20
- We fine tune this model to evaluate (regression) the clickbait level of title news.
21
 
22
  ## Intended uses & limitations
23
 
24
- Model looks like the model described in the paper [Predicting Clickbait Strength in Online Social Media](https://aclanthology.org/2020.coling-main.425/) by Indurthi Vijayasaradhi, Syed Bakhtiyar, Gupta Manish, Varma Vasudeva.
25
-
26
- The model was trained with english titles.
27
 
28
  ## Training and evaluation data
29
 
30
- We trained the model with the official training data for the chalenge (clickbait17-train-170630.zip (894 MiB, 19538 posts), plus another set that was just available after the end of the challenge (clickbait17-train-170331.zip (157 MiB, 2459 posts).
31
 
32
  ## Training procedure
33
- Code can be find in [Github](https://github.com/caush/Clickbait).
34
 
35
  ### Training hyperparameters
36
 
@@ -47,51 +45,60 @@ The following hyperparameters were used during training:
47
 
48
  | Training Loss | Epoch | Step | Validation Loss |
49
  |:-------------:|:-----:|:----:|:---------------:|
50
- | No log | 0.05 | 50 | 0.0390 |
51
- | No log | 0.09 | 100 | 0.0365 |
52
- | No log | 0.14 | 150 | 0.0321 |
53
- | No log | 0.18 | 200 | 0.0319 |
54
- | No log | 0.23 | 250 | 0.0350 |
55
- | No log | 0.27 | 300 | 0.0317 |
56
- | No log | 0.32 | 350 | 0.0304 |
57
- | No log | 0.36 | 400 | 0.0283 |
58
- | No log | 0.41 | 450 | 0.0300 |
59
- | 0.0364 | 0.46 | 500 | 0.0282 |
60
- | 0.0364 | 0.5 | 550 | 0.0285 |
61
- | 0.0364 | 0.55 | 600 | 0.0276 |
62
- | 0.0364 | 0.59 | 650 | 0.0278 |
63
- | 0.0364 | 0.64 | 700 | 0.0293 |
64
- | 0.0364 | 0.68 | 750 | 0.0280 |
65
- | 0.0364 | 0.73 | 800 | 0.0320 |
66
- | 0.0364 | 0.77 | 850 | 0.0269 |
67
- | 0.0364 | 0.82 | 900 | 0.0269 |
68
- | 0.0364 | 0.87 | 950 | 0.0271 |
69
- | 0.0281 | 0.91 | 1000 | 0.0314 |
70
- | 0.0281 | 0.96 | 1050 | 0.0265 |
71
- | 0.0281 | 1.0 | 1100 | 0.0295 |
72
- | 0.0281 | 1.05 | 1150 | 0.0295 |
73
- | 0.0281 | 1.09 | 1200 | 0.0290 |
74
- | 0.0281 | 1.14 | 1250 | 0.0281 |
75
- | 0.0281 | 1.18 | 1300 | 0.0272 |
76
- | 0.0281 | 1.23 | 1350 | 0.0273 |
77
- | 0.0281 | 1.28 | 1400 | 0.0287 |
78
- | 0.0281 | 1.32 | 1450 | 0.0267 |
79
- | 0.026 | 1.37 | 1500 | 0.0284 |
80
- | 0.026 | 1.41 | 1550 | 0.0264 |
81
- | 0.026 | 1.46 | 1600 | 0.0273 |
82
- | 0.026 | 1.5 | 1650 | 0.0280 |
83
- | 0.026 | 1.55 | 1700 | 0.0266 |
84
- | 0.026 | 1.59 | 1750 | 0.0260 |
85
- | 0.026 | 1.64 | 1800 | 0.0266 |
86
- | 0.026 | 1.68 | 1850 | 0.0268 |
87
- | 0.026 | 1.73 | 1900 | 0.0269 |
88
- | 0.026 | 1.78 | 1950 | 0.0260 |
89
- | 0.0236 | 1.82 | 2000 | 0.0273 |
90
- | 0.0236 | 1.87 | 2050 | 0.0272 |
91
- | 0.0236 | 1.91 | 2100 | 0.0260 |
92
- | 0.0236 | 1.96 | 2150 | 0.0269 |
93
- | 0.0236 | 2.0 | 2200 | 0.0286 |
94
- | 0.0236 | 2.05 | 2250 | 0.0266 |
 
 
 
 
 
 
 
 
 
95
 
96
 
97
  ### Framework versions
 
7
  results: []
8
  ---
9
 
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
  # Clickbait1
14
 
15
+ This model is a fine-tuned version of [microsoft/Multilingual-MiniLM-L12-H384](https://huggingface.co/microsoft/Multilingual-MiniLM-L12-H384) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.0257
18
 
19
  ## Model description
20
 
21
+ More information needed
 
 
22
 
23
  ## Intended uses & limitations
24
 
25
+ More information needed
 
 
26
 
27
  ## Training and evaluation data
28
 
29
+ More information needed
30
 
31
  ## Training procedure
 
32
 
33
  ### Training hyperparameters
34
 
 
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
+ | No log | 0.05 | 50 | 0.0571 |
49
+ | No log | 0.09 | 100 | 0.0448 |
50
+ | No log | 0.14 | 150 | 0.0391 |
51
+ | No log | 0.18 | 200 | 0.0326 |
52
+ | No log | 0.23 | 250 | 0.0343 |
53
+ | No log | 0.27 | 300 | 0.0343 |
54
+ | No log | 0.32 | 350 | 0.0343 |
55
+ | No log | 0.36 | 400 | 0.0346 |
56
+ | No log | 0.41 | 450 | 0.0343 |
57
+ | 0.0388 | 0.46 | 500 | 0.0297 |
58
+ | 0.0388 | 0.5 | 550 | 0.0293 |
59
+ | 0.0388 | 0.55 | 600 | 0.0301 |
60
+ | 0.0388 | 0.59 | 650 | 0.0290 |
61
+ | 0.0388 | 0.64 | 700 | 0.0326 |
62
+ | 0.0388 | 0.68 | 750 | 0.0285 |
63
+ | 0.0388 | 0.73 | 800 | 0.0285 |
64
+ | 0.0388 | 0.77 | 850 | 0.0275 |
65
+ | 0.0388 | 0.82 | 900 | 0.0314 |
66
+ | 0.0388 | 0.87 | 950 | 0.0309 |
67
+ | 0.0297 | 0.91 | 1000 | 0.0277 |
68
+ | 0.0297 | 0.96 | 1050 | 0.0281 |
69
+ | 0.0297 | 1.0 | 1100 | 0.0273 |
70
+ | 0.0297 | 1.05 | 1150 | 0.0270 |
71
+ | 0.0297 | 1.09 | 1200 | 0.0291 |
72
+ | 0.0297 | 1.14 | 1250 | 0.0293 |
73
+ | 0.0297 | 1.18 | 1300 | 0.0269 |
74
+ | 0.0297 | 1.23 | 1350 | 0.0276 |
75
+ | 0.0297 | 1.28 | 1400 | 0.0279 |
76
+ | 0.0297 | 1.32 | 1450 | 0.0267 |
77
+ | 0.0265 | 1.37 | 1500 | 0.0270 |
78
+ | 0.0265 | 1.41 | 1550 | 0.0300 |
79
+ | 0.0265 | 1.46 | 1600 | 0.0274 |
80
+ | 0.0265 | 1.5 | 1650 | 0.0274 |
81
+ | 0.0265 | 1.55 | 1700 | 0.0266 |
82
+ | 0.0265 | 1.59 | 1750 | 0.0267 |
83
+ | 0.0265 | 1.64 | 1800 | 0.0267 |
84
+ | 0.0265 | 1.68 | 1850 | 0.0280 |
85
+ | 0.0265 | 1.73 | 1900 | 0.0274 |
86
+ | 0.0265 | 1.78 | 1950 | 0.0272 |
87
+ | 0.025 | 1.82 | 2000 | 0.0261 |
88
+ | 0.025 | 1.87 | 2050 | 0.0268 |
89
+ | 0.025 | 1.91 | 2100 | 0.0268 |
90
+ | 0.025 | 1.96 | 2150 | 0.0259 |
91
+ | 0.025 | 2.0 | 2200 | 0.0257 |
92
+ | 0.025 | 2.05 | 2250 | 0.0260 |
93
+ | 0.025 | 2.09 | 2300 | 0.0263 |
94
+ | 0.025 | 2.14 | 2350 | 0.0262 |
95
+ | 0.025 | 2.19 | 2400 | 0.0269 |
96
+ | 0.025 | 2.23 | 2450 | 0.0262 |
97
+ | 0.0223 | 2.28 | 2500 | 0.0262 |
98
+ | 0.0223 | 2.32 | 2550 | 0.0267 |
99
+ | 0.0223 | 2.37 | 2600 | 0.0260 |
100
+ | 0.0223 | 2.41 | 2650 | 0.0260 |
101
+ | 0.0223 | 2.46 | 2700 | 0.0259 |
102
 
103
 
104
  ### Framework versions