Hugging Face's logo

YagiASAFAS
/
MSPoliBERT-12

PyTorch
bert
Generated from Trainer
Model card Files
xet
Community
YagiASAFAS commited on
Commit
312bdcd
·
1 Parent(s): d835ccd

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +92 -3
README.md CHANGED
@@ -1,3 +1,92 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: MSPoliBERT-12
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
+ # MSPoliBERT-12
14
+
15
+ This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.2936
18
+ - Democracy F1: 0.9392
19
+ - Democracy Accuracy: 0.9426
20
+ - Economy F1: 0.9141
21
+ - Economy Accuracy: 0.9156
22
+ - Race F1: 0.9303
23
+ - Race Accuracy: 0.9331
24
+ - Leadership F1: 0.7696
25
+ - Leadership Accuracy: 0.7688
26
+ - Development F1: 0.8747
27
+ - Development Accuracy: 0.8790
28
+ - Corruption F1: 0.9411
29
+ - Corruption Accuracy: 0.9441
30
+ - Instability F1: 0.9093
31
+ - Instability Accuracy: 0.9141
32
+ - Safety F1: 0.9291
33
+ - Safety Accuracy: 0.9305
34
+ - Administration F1: 0.8768
35
+ - Administration Accuracy: 0.8853
36
+ - Education F1: 0.9538
37
+ - Education Accuracy: 0.9557
38
+ - Religion F1: 0.9338
39
+ - Religion Accuracy: 0.9349
40
+ - Environment F1: 0.9807
41
+ - Environment Accuracy: 0.9819
42
+ - Overall F1: 0.9127
43
+ - Overall Accuracy: 0.9155
44
+
45
+ ## Model description
46
+
47
+ More information needed
48
+
49
+ ## Intended uses & limitations
50
+
51
+ More information needed
52
+
53
+ ## Training and evaluation data
54
+
55
+ More information needed
56
+
57
+ ## Training procedure
58
+
59
+ ### Training hyperparameters
60
+
61
+ The following hyperparameters were used during training:
62
+ - learning_rate: 5e-05
63
+ - train_batch_size: 8
64
+ - eval_batch_size: 8
65
+ - seed: 42
66
+ - gradient_accumulation_steps: 2
67
+ - total_train_batch_size: 16
68
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
69
+ - lr_scheduler_type: linear
70
+ - num_epochs: 8
71
+ - mixed_precision_training: Native AMP
72
+
73
+ ### Training results
74
+
75
+ | Training Loss | Epoch | Step | Validation Loss | Democracy F1 | Democracy Accuracy | Economy F1 | Economy Accuracy | Race F1 | Race Accuracy | Leadership F1 | Leadership Accuracy | Development F1 | Development Accuracy | Corruption F1 | Corruption Accuracy | Instability F1 | Instability Accuracy | Safety F1 | Safety Accuracy | Administration F1 | Administration Accuracy | Education F1 | Education Accuracy | Religion F1 | Religion Accuracy | Environment F1 | Environment Accuracy | Overall F1 | Overall Accuracy |
76
+ |:-------------:|:-----:|:----:|:---------------:|:------------:|:------------------:|:----------:|:----------------:|:-------:|:-------------:|:-------------:|:-------------------:|:--------------:|:--------------------:|:-------------:|:-------------------:|:--------------:|:--------------------:|:---------:|:---------------:|:-----------------:|:-----------------------:|:------------:|:------------------:|:-----------:|:-----------------:|:--------------:|:--------------------:|:----------:|:----------------:|
77
+ | 0.4282 | 1.0 | 841 | 0.2914 | 0.9080 | 0.9293 | 0.8960 | 0.9088 | 0.9066 | 0.9221 | 0.7142 | 0.7328 | 0.8409 | 0.8585 | 0.9253 | 0.9287 | 0.9013 | 0.9076 | 0.9076 | 0.9097 | 0.8349 | 0.8651 | 0.9376 | 0.9483 | 0.9147 | 0.9233 | 0.9671 | 0.9744 | 0.8878 | 0.9007 |
78
+ | 0.2346 | 2.0 | 1682 | 0.2568 | 0.9172 | 0.9364 | 0.9016 | 0.9105 | 0.9172 | 0.9254 | 0.7547 | 0.7652 | 0.8586 | 0.8648 | 0.9265 | 0.9346 | 0.8974 | 0.9111 | 0.9272 | 0.9296 | 0.8539 | 0.8802 | 0.9451 | 0.9519 | 0.9264 | 0.9308 | 0.9767 | 0.9786 | 0.9002 | 0.9099 |
79
+ | 0.1601 | 3.0 | 2523 | 0.2519 | 0.9260 | 0.9355 | 0.9108 | 0.9186 | 0.9228 | 0.9278 | 0.7575 | 0.7620 | 0.8748 | 0.8808 | 0.9360 | 0.9415 | 0.9067 | 0.9135 | 0.9285 | 0.9316 | 0.8609 | 0.8799 | 0.9518 | 0.9560 | 0.9301 | 0.9337 | 0.9801 | 0.9810 | 0.9072 | 0.9135 |
80
+ | 0.1169 | 4.0 | 3364 | 0.2627 | 0.9315 | 0.9412 | 0.9120 | 0.9192 | 0.9214 | 0.9284 | 0.7637 | 0.7646 | 0.8757 | 0.8799 | 0.9411 | 0.9459 | 0.9071 | 0.9123 | 0.9296 | 0.9328 | 0.8685 | 0.8820 | 0.9512 | 0.9542 | 0.9335 | 0.9364 | 0.9802 | 0.9810 | 0.9096 | 0.9148 |
81
+ | 0.0798 | 5.0 | 4205 | 0.2729 | 0.9368 | 0.9412 | 0.9129 | 0.9159 | 0.9284 | 0.9328 | 0.7642 | 0.7652 | 0.8760 | 0.8799 | 0.9414 | 0.9435 | 0.9078 | 0.9126 | 0.9277 | 0.9290 | 0.8703 | 0.8743 | 0.9565 | 0.9581 | 0.9323 | 0.9349 | 0.9799 | 0.9801 | 0.9112 | 0.9140 |
82
+ | 0.0565 | 6.0 | 5046 | 0.2821 | 0.9357 | 0.9403 | 0.9144 | 0.9159 | 0.9266 | 0.9284 | 0.7687 | 0.7685 | 0.8748 | 0.8785 | 0.9384 | 0.9403 | 0.9115 | 0.9153 | 0.9266 | 0.9299 | 0.8693 | 0.8814 | 0.9557 | 0.9578 | 0.9321 | 0.9325 | 0.9790 | 0.9813 | 0.9111 | 0.9142 |
83
+ | 0.0443 | 7.0 | 5887 | 0.2914 | 0.9375 | 0.9406 | 0.9150 | 0.9156 | 0.9293 | 0.9322 | 0.7719 | 0.7715 | 0.8727 | 0.8767 | 0.9412 | 0.9447 | 0.9103 | 0.9144 | 0.9292 | 0.9316 | 0.8761 | 0.8832 | 0.9558 | 0.9569 | 0.9322 | 0.9334 | 0.9797 | 0.9813 | 0.9126 | 0.9152 |
84
+ | 0.0361 | 8.0 | 6728 | 0.2936 | 0.9392 | 0.9426 | 0.9141 | 0.9156 | 0.9303 | 0.9331 | 0.7696 | 0.7688 | 0.8747 | 0.8790 | 0.9411 | 0.9441 | 0.9093 | 0.9141 | 0.9291 | 0.9305 | 0.8768 | 0.8853 | 0.9538 | 0.9557 | 0.9338 | 0.9349 | 0.9807 | 0.9819 | 0.9127 | 0.9155 |
85
+
86
+
87
+ ### Framework versions
88
+
89
+ - Transformers 4.18.0
90
+ - Pytorch 2.5.0+cu121
91
+ - Datasets 3.1.0
92
+ - Tokenizers 0.12.1