Hugging Face's logo

YagiASAFAS
/
MyPoliBERT-ver02

Text Classification
Transformers
PyTorch
Safetensors
bert
Generated from Trainer
Model card Files Community
YagiASAFAS commited on
Commit
cdd7a43
·
1 Parent(s): 939412c

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +93 -3
README.md CHANGED
@@ -1,3 +1,93 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: MyPoliBERT-ver02
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
+ # MyPoliBERT-ver02
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: 0.2953
18
+ - Democracy F1: 0.9401
19
+ - Democracy Accuracy: 0.9392
20
+ - Economy F1: 0.9191
21
+ - Economy Accuracy: 0.9182
22
+ - Race F1: 0.9521
23
+ - Race Accuracy: 0.9516
24
+ - Leadership F1: 0.8198
25
+ - Leadership Accuracy: 0.8196
26
+ - Development F1: 0.8877
27
+ - Development Accuracy: 0.8869
28
+ - Corruption F1: 0.9487
29
+ - Corruption Accuracy: 0.9498
30
+ - Instability F1: 0.9254
31
+ - Instability Accuracy: 0.9283
32
+ - Safety F1: 0.9209
33
+ - Safety Accuracy: 0.9207
34
+ - Administration F1: 0.8993
35
+ - Administration Accuracy: 0.9019
36
+ - Education F1: 0.9632
37
+ - Education Accuracy: 0.9632
38
+ - Religion F1: 0.9557
39
+ - Religion Accuracy: 0.9554
40
+ - Environment F1: 0.9734
41
+ - Environment Accuracy: 0.9729
42
+ - Overall F1: 0.9255
43
+ - Overall Accuracy: 0.9256
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: 16
64
+ - eval_batch_size: 16
65
+ - seed: 42
66
+ - gradient_accumulation_steps: 2
67
+ - total_train_batch_size: 32
68
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
69
+ - lr_scheduler_type: linear
70
+ - num_epochs: 16
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
+ | 0.4126 | 1.0 | 757 | 0.2492 | 0.9175 | 0.9283 | 0.9010 | 0.9096 | 0.9374 | 0.9433 | 0.7897 | 0.7896 | 0.8463 | 0.8655 | 0.9349 | 0.9422 | 0.9088 | 0.9201 | 0.9165 | 0.9199 | 0.8760 | 0.8918 | 0.9546 | 0.9592 | 0.9480 | 0.9493 | 0.9704 | 0.9713 | 0.9084 | 0.9158 |
78
+ | 0.2014 | 2.0 | 1514 | 0.2217 | 0.9346 | 0.9389 | 0.9180 | 0.9191 | 0.9508 | 0.9528 | 0.8148 | 0.8195 | 0.8814 | 0.8905 | 0.9433 | 0.9427 | 0.9133 | 0.9106 | 0.9155 | 0.9153 | 0.8913 | 0.9052 | 0.9641 | 0.9658 | 0.9513 | 0.9519 | 0.9723 | 0.9741 | 0.9209 | 0.9239 |
79
+ | 0.1422 | 3.0 | 2271 | 0.2244 | 0.9369 | 0.9397 | 0.9202 | 0.9224 | 0.9503 | 0.9500 | 0.8140 | 0.8188 | 0.8824 | 0.8888 | 0.9453 | 0.9447 | 0.9227 | 0.9250 | 0.9182 | 0.9172 | 0.8926 | 0.9014 | 0.9646 | 0.9658 | 0.9529 | 0.9549 | 0.9745 | 0.9746 | 0.9229 | 0.9253 |
80
+ | 0.098 | 4.0 | 3028 | 0.2310 | 0.9373 | 0.9405 | 0.9239 | 0.9257 | 0.9532 | 0.9541 | 0.8112 | 0.8142 | 0.8894 | 0.8941 | 0.9470 | 0.9466 | 0.9233 | 0.9243 | 0.9180 | 0.9196 | 0.8967 | 0.9068 | 0.9619 | 0.9630 | 0.9527 | 0.9534 | 0.9742 | 0.9751 | 0.9241 | 0.9265 |
81
+ | 0.0722 | 5.0 | 3785 | 0.2507 | 0.9368 | 0.9379 | 0.9254 | 0.9277 | 0.9531 | 0.9536 | 0.8115 | 0.8117 | 0.8862 | 0.8880 | 0.9424 | 0.9405 | 0.9262 | 0.9285 | 0.9139 | 0.9126 | 0.8961 | 0.8987 | 0.9631 | 0.9642 | 0.9548 | 0.9552 | 0.9750 | 0.9746 | 0.9237 | 0.9244 |
82
+ | 0.053 | 6.0 | 4542 | 0.2619 | 0.9405 | 0.9424 | 0.9216 | 0.9220 | 0.9536 | 0.9546 | 0.8155 | 0.8132 | 0.8877 | 0.8907 | 0.9522 | 0.9542 | 0.9215 | 0.9179 | 0.9210 | 0.9207 | 0.8976 | 0.8999 | 0.9611 | 0.9605 | 0.9547 | 0.9544 | 0.9750 | 0.9751 | 0.9252 | 0.9255 |
83
+ | 0.0413 | 7.0 | 5299 | 0.2727 | 0.9414 | 0.9440 | 0.9236 | 0.9235 | 0.9534 | 0.9541 | 0.8247 | 0.8244 | 0.8869 | 0.8890 | 0.9491 | 0.9495 | 0.9264 | 0.9275 | 0.9224 | 0.9227 | 0.8977 | 0.9057 | 0.9635 | 0.9642 | 0.9527 | 0.9523 | 0.9738 | 0.9739 | 0.9263 | 0.9276 |
84
+ | 0.0322 | 8.0 | 6056 | 0.2880 | 0.9389 | 0.9410 | 0.9198 | 0.9234 | 0.9544 | 0.9547 | 0.8142 | 0.8099 | 0.8872 | 0.8878 | 0.9522 | 0.9549 | 0.9274 | 0.9288 | 0.9208 | 0.9214 | 0.9027 | 0.9068 | 0.9632 | 0.9640 | 0.9534 | 0.9536 | 0.9745 | 0.9744 | 0.9257 | 0.9267 |
85
+ | 0.0256 | 9.0 | 6813 | 0.2953 | 0.9401 | 0.9392 | 0.9191 | 0.9182 | 0.9521 | 0.9516 | 0.8198 | 0.8196 | 0.8877 | 0.8869 | 0.9487 | 0.9498 | 0.9254 | 0.9283 | 0.9209 | 0.9207 | 0.8993 | 0.9019 | 0.9632 | 0.9632 | 0.9557 | 0.9554 | 0.9734 | 0.9729 | 0.9255 | 0.9256 |
86
+
87
+
88
+ ### Framework versions
89
+
90
+ - Transformers 4.18.0
91
+ - Pytorch 2.5.1+cu121
92
+ - Datasets 3.2.0
93
+ - Tokenizers 0.12.1