Hugging Face's logo

Climate-TwitterBERT
/
Climate-TwitterBERT-step2

Text Classification
Transformers
PyTorch
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
Climate-TwitterBERT commited on
Commit
d68f4f1
·
1 Parent(s): 491319c

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +83 -0
README.md ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ - precision
8
+ - recall
9
+ - f1
10
+ model-index:
11
+ - name: Climate-TwitterBERT-xmas
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # Climate-TwitterBERT-xmas
19
+
20
+ This model is a fine-tuned version of [digitalepidemiologylab/covid-twitter-bert-v2](https://huggingface.co/digitalepidemiologylab/covid-twitter-bert-v2) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.3488
23
+ - Accuracy: 0.9
24
+ - Precision: 0.7963
25
+ - Recall: 0.7544
26
+ - F1-weighted: 0.8990
27
+ - F1: 0.7748
28
+
29
+ ## Model description
30
+
31
+ More information needed
32
+
33
+ ## Intended uses & limitations
34
+
35
+ More information needed
36
+
37
+ ## Training and evaluation data
38
+
39
+ More information needed
40
+
41
+ ## Training procedure
42
+
43
+ ### Training hyperparameters
44
+
45
+ The following hyperparameters were used during training:
46
+ - learning_rate: 1e-05
47
+ - train_batch_size: 8
48
+ - eval_batch_size: 2
49
+ - seed: 42
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: linear
52
+ - lr_scheduler_warmup_ratio: 0.05
53
+ - num_epochs: 4
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1-weighted | F1 |
58
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:-----------:|:------:|
59
+ | 0.5977 | 0.23 | 50 | 0.5488 | 0.772 | 0.5 | 0.0351 | 0.6867 | 0.0656 |
60
+ | 0.5246 | 0.46 | 100 | 0.4535 | 0.804 | 0.6053 | 0.4035 | 0.7890 | 0.4842 |
61
+ | 0.461 | 0.68 | 150 | 0.4056 | 0.832 | 0.8261 | 0.3333 | 0.8031 | 0.475 |
62
+ | 0.4096 | 0.91 | 200 | 0.3367 | 0.86 | 0.6897 | 0.7018 | 0.8604 | 0.6957 |
63
+ | 0.3217 | 1.14 | 250 | 0.3402 | 0.852 | 0.6562 | 0.7368 | 0.8549 | 0.6942 |
64
+ | 0.2896 | 1.37 | 300 | 0.3189 | 0.86 | 0.6964 | 0.6842 | 0.8596 | 0.6903 |
65
+ | 0.3433 | 1.6 | 350 | 0.3217 | 0.888 | 0.8372 | 0.6316 | 0.8821 | 0.72 |
66
+ | 0.3306 | 1.83 | 400 | 0.3117 | 0.876 | 0.7241 | 0.7368 | 0.8764 | 0.7304 |
67
+ | 0.2853 | 2.05 | 450 | 0.2913 | 0.876 | 0.7167 | 0.7544 | 0.8771 | 0.7350 |
68
+ | 0.2066 | 2.28 | 500 | 0.2879 | 0.904 | 0.8 | 0.7719 | 0.9034 | 0.7857 |
69
+ | 0.2382 | 2.51 | 550 | 0.2963 | 0.9 | 0.7857 | 0.7719 | 0.8997 | 0.7788 |
70
+ | 0.2293 | 2.74 | 600 | 0.3065 | 0.9 | 0.8478 | 0.6842 | 0.8960 | 0.7573 |
71
+ | 0.287 | 2.97 | 650 | 0.3184 | 0.912 | 0.8431 | 0.7544 | 0.9102 | 0.7963 |
72
+ | 0.1818 | 3.2 | 700 | 0.3442 | 0.912 | 0.8431 | 0.7544 | 0.9102 | 0.7963 |
73
+ | 0.186 | 3.42 | 750 | 0.3435 | 0.912 | 0.8431 | 0.7544 | 0.9102 | 0.7963 |
74
+ | 0.1409 | 3.65 | 800 | 0.3477 | 0.904 | 0.8113 | 0.7544 | 0.9027 | 0.7818 |
75
+ | 0.2049 | 3.88 | 850 | 0.3488 | 0.9 | 0.7963 | 0.7544 | 0.8990 | 0.7748 |
76
+
77
+
78
+ ### Framework versions
79
+
80
+ - Transformers 4.28.1
81
+ - Pytorch 2.1.0+cu121
82
+ - Datasets 2.16.0
83
+ - Tokenizers 0.13.3