Hugging Face's logo

thejosango
/
nuha-binary

Text Classification
Transformers
PyTorch
Arabic
bert
hate-speech
gender-based-violence
arabic
binary-classification
pilot
Eval Results (legacy)
text-embeddings-inference
Model card Files
xet
 Community
yazansh commited on
Commit
714ad98
·
verified ·
1 Parent(s): cc3ed62

binary-26

Browse files
Files changed (4) hide show
  1. README.md +33 -31
  2. config.toml +2 -2
  3. pytorch_model.bin +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -23,13 +23,13 @@ model-index:
23
  metrics:
24
  - name: F1
25
  type: f1
26
- value: 0.6711851987543506
27
  - name: Precision
28
  type: precision
29
- value: 0.6655767484105358
30
  - name: Recall
31
  type: recall
32
- value: 0.676888970995751
33
  ---
34
 
35
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -39,10 +39,10 @@ should probably proofread and complete it, then remove this comment. -->
39
 
40
  This model is a fine-tuned version of [thejosango/nuha-mlm](https://huggingface.co/thejosango/nuha-mlm) on the nuha-dataset dataset.
41
  It achieves the following results on the evaluation set:
42
- - Loss: 0.5283
43
- - F1: 0.6712
44
- - Precision: 0.6656
45
- - Recall: 0.6769
46
  - Support: None
47
 
48
  ## Model description
@@ -78,30 +78,32 @@ The following hyperparameters were used during training:
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
80
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:---------:|:------:|:-------:|
81
- | 1.2633 | 0.3 | 500 | 0.6344 | 0.5719 | 0.5563 | 0.5884 | None |
82
- | 0.9118 | 0.6 | 1000 | 0.5886 | 0.5940 | 0.5973 | 0.5908 | None |
83
- | 0.7827 | 0.9 | 1500 | 0.5890 | 0.6367 | 0.5640 | 0.7308 | None |
84
- | 0.6804 | 1.2 | 2000 | 0.5905 | 0.6458 | 0.5495 | 0.7829 | None |
85
- | 0.6492 | 1.49 | 2500 | 0.5619 | 0.6558 | 0.5993 | 0.7242 | None |
86
- | 0.6268 | 1.79 | 3000 | 0.5676 | 0.6642 | 0.5844 | 0.7691 | None |
87
- | 0.6148 | 2.09 | 3500 | 0.5476 | 0.6510 | 0.6160 | 0.6902 | None |
88
- | 0.5816 | 2.39 | 4000 | 0.5492 | 0.6666 | 0.6142 | 0.7286 | None |
89
- | 0.5855 | 2.69 | 4500 | 0.5549 | 0.6716 | 0.6043 | 0.7560 | None |
90
- | 0.5712 | 2.99 | 5000 | 0.5285 | 0.6665 | 0.6445 | 0.6900 | None |
91
- | 0.5507 | 3.29 | 5500 | 0.5435 | 0.6705 | 0.6187 | 0.7318 | None |
92
- | 0.5543 | 3.59 | 6000 | 0.5324 | 0.6772 | 0.6387 | 0.7207 | None |
93
- | 0.5448 | 3.89 | 6500 | 0.5254 | 0.6711 | 0.6555 | 0.6874 | None |
94
- | 0.5313 | 4.18 | 7000 | 0.5428 | 0.6823 | 0.6219 | 0.7558 | None |
95
- | 0.5268 | 4.48 | 7500 | 0.5192 | 0.6667 | 0.6758 | 0.6579 | None |
96
- | 0.5242 | 4.78 | 8000 | 0.5330 | 0.6844 | 0.6360 | 0.7406 | None |
97
- | 0.519 | 5.08 | 8500 | 0.5203 | 0.6650 | 0.6837 | 0.6473 | None |
98
- | 0.5056 | 5.38 | 9000 | 0.5607 | 0.6865 | 0.6080 | 0.7881 | None |
99
- | 0.5025 | 5.68 | 9500 | 0.5238 | 0.6429 | 0.7203 | 0.5805 | None |
100
- | 0.5086 | 5.98 | 10000 | 0.6008 | 0.6791 | 0.5661 | 0.8485 | None |
101
- | 0.4832 | 6.28 | 10500 | 0.5555 | 0.6892 | 0.6189 | 0.7776 | None |
102
- | 0.494 | 6.58 | 11000 | 0.5286 | 0.6773 | 0.6538 | 0.7026 | None |
103
- | 0.4849 | 6.87 | 11500 | 0.5570 | 0.6867 | 0.6154 | 0.7766 | None |
104
- | 0.4743 | 7.17 | 12000 | 0.5283 | 0.6712 | 0.6656 | 0.6769 | None |
 
 
105
 
106
 
107
  ### Framework versions
 
23
  metrics:
24
  - name: F1
25
  type: f1
26
+ value: 0.6709883502442691
27
  - name: Precision
28
  type: precision
29
+ value: 0.6826610590709233
30
  - name: Recall
31
  type: recall
32
+ value: 0.6597081101053021
33
  ---
34
 
35
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
39
 
40
  This model is a fine-tuned version of [thejosango/nuha-mlm](https://huggingface.co/thejosango/nuha-mlm) on the nuha-dataset dataset.
41
  It achieves the following results on the evaluation set:
42
+ - Loss: 0.5203
43
+ - F1: 0.6710
44
+ - Precision: 0.6827
45
+ - Recall: 0.6597
46
  - Support: None
47
 
48
  ## Model description
 
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
80
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:---------:|:------:|:-------:|
81
+ | 1.2252 | 0.23 | 500 | 0.6446 | 0.5213 | 0.6006 | 0.4606 | None |
82
+ | 0.8851 | 0.46 | 1000 | 0.5788 | 0.6011 | 0.6123 | 0.5904 | None |
83
+ | 0.7632 | 0.7 | 1500 | 0.5560 | 0.5957 | 0.6485 | 0.5509 | None |
84
+ | 0.684 | 0.93 | 2000 | 0.5481 | 0.6076 | 0.6506 | 0.5699 | None |
85
+ | 0.6334 | 1.16 | 2500 | 0.6265 | 0.6462 | 0.5199 | 0.8535 | None |
86
+ | 0.6137 | 1.39 | 3000 | 0.5378 | 0.5593 | 0.7224 | 0.4563 | None |
87
+ | 0.609 | 1.63 | 3500 | 0.5348 | 0.6486 | 0.6499 | 0.6473 | None |
88
+ | 0.5756 | 1.86 | 4000 | 0.5272 | 0.6182 | 0.6966 | 0.5557 | None |
89
+ | 0.5667 | 2.09 | 4500 | 0.5257 | 0.6035 | 0.7194 | 0.5197 | None |
90
+ | 0.5551 | 2.32 | 5000 | 0.5193 | 0.6456 | 0.6837 | 0.6115 | None |
91
+ | 0.5549 | 2.56 | 5500 | 0.5173 | 0.6236 | 0.7191 | 0.5505 | None |
92
+ | 0.5507 | 2.79 | 6000 | 0.5279 | 0.6675 | 0.6542 | 0.6813 | None |
93
+ | 0.5505 | 3.02 | 6500 | 0.5164 | 0.6534 | 0.6916 | 0.6192 | None |
94
+ | 0.5298 | 3.25 | 7000 | 0.5232 | 0.6687 | 0.6628 | 0.6747 | None |
95
+ | 0.5313 | 3.49 | 7500 | 0.5128 | 0.6603 | 0.6934 | 0.6303 | None |
96
+ | 0.5209 | 3.72 | 8000 | 0.5285 | 0.6800 | 0.6513 | 0.7113 | None |
97
+ | 0.5214 | 3.95 | 8500 | 0.5127 | 0.6443 | 0.7128 | 0.5878 | None |
98
+ | 0.5033 | 4.18 | 9000 | 0.5179 | 0.6341 | 0.7268 | 0.5623 | None |
99
+ | 0.5055 | 4.41 | 9500 | 0.5214 | 0.6239 | 0.7347 | 0.5422 | None |
100
+ | 0.5013 | 4.65 | 10000 | 0.5230 | 0.6794 | 0.6626 | 0.6972 | None |
101
+ | 0.5107 | 4.88 | 10500 | 0.5127 | 0.6656 | 0.7012 | 0.6335 | None |
102
+ | 0.4862 | 5.11 | 11000 | 0.5447 | 0.5848 | 0.7670 | 0.4726 | None |
103
+ | 0.4814 | 5.34 | 11500 | 0.5216 | 0.6217 | 0.7386 | 0.5367 | None |
104
+ | 0.4877 | 5.58 | 12000 | 0.5176 | 0.6375 | 0.7255 | 0.5684 | None |
105
+ | 0.4893 | 5.81 | 12500 | 0.5180 | 0.6463 | 0.7129 | 0.5912 | None |
106
+ | 0.4779 | 6.04 | 13000 | 0.5203 | 0.6710 | 0.6827 | 0.6597 | None |
107
 
108
 
109
  ### Framework versions
config.toml CHANGED
@@ -1,5 +1,5 @@
1
  [experiment]
2
- name = "binary-25"
3
  type = "binary"
4
 
5
 
@@ -7,7 +7,7 @@ type = "binary"
7
  path = "thejosango/nuha-dataset"
8
  dataset_revision = "main"
9
  augment_ratio = 0.25
10
- undersampling_strategy = "majority"
11
 
12
 
13
  [model]
 
1
  [experiment]
2
+ name = "binary-26"
3
  type = "binary"
4
 
5
 
 
7
  path = "thejosango/nuha-dataset"
8
  dataset_revision = "main"
9
  augment_ratio = 0.25
10
+ undersampling_strategy = 0
11
 
12
 
13
  [model]
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e7a98781949be70d56a87a882c2f59124d9d2df230e6f29e72b575c7082c3f0a
3
  size 313992076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6274d0bde584699734a00b2cd752b54dcaf69e5a4753cab991977f54ca510145
3
  size 313992076
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5000e8929849a0d1fc06cf7a92f266db77a62a36dee5f486357cad977bf553ac
3
  size 4091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cc61a0a4012cb94dd109030c3f0bc5d60b87d98e45eac45d39704943b4e479e
3
  size 4091