Hugging Face's logo

thejosango
/
nuha-ajp-binary

Text Classification
Transformers
PyTorch
Safetensors
Arabic
bert
hate-speech
gender-based-violence
arabic
binary-classification
pilot
Eval Results (legacy)
text-embeddings-inference
Model card Files
xet
 Community
4
yazansh commited on
Commit
c79de0c
·
verified ·
1 Parent(s): 0636783

binary-32

Browse files
Files changed (5) hide show
  1. README.md +18 -18
  2. config.toml +2 -2
  3. pytorch_model.bin +1 -1
  4. tokenizer.json +8 -1
  5. training_args.bin +1 -1
README.md CHANGED
@@ -23,13 +23,13 @@ model-index:
23
  metrics:
24
  - name: F1
25
  type: f1
26
- value: 0.6931693758948876
27
  - name: Precision
28
  type: precision
29
- value: 0.6369969040247678
30
  - name: Recall
31
  type: recall
32
- value: 0.7602069092924442
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: 0.5323
43
- - F1: 0.6932
44
- - Precision: 0.6370
45
- - Recall: 0.7602
46
  - Support: None
47
 
48
  ## Model description
@@ -69,7 +69,7 @@ The following hyperparameters were used during training:
69
  - gradient_accumulation_steps: 2
70
  - total_train_batch_size: 64
71
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
72
- - lr_scheduler_type: constant
73
  - num_epochs: 30
74
  - label_smoothing_factor: 0.1
75
 
@@ -77,16 +77,16 @@ The following hyperparameters were used during training:
77
 
78
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
79
  |:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:|:-------:|
80
- | 0.5818 | 0.37 | 500 | 0.6018 | 0.6494 | 0.5313 | 0.8352 | None |
81
- | 0.5609 | 0.74 | 1000 | 0.5822 | 0.6641 | 0.5548 | 0.8273 | None |
82
- | 0.5464 | 1.11 | 1500 | 0.5574 | 0.6743 | 0.5874 | 0.7914 | None |
83
- | 0.534 | 1.48 | 2000 | 0.5228 | 0.6766 | 0.6473 | 0.7087 | None |
84
- | 0.5293 | 1.85 | 2500 | 0.5279 | 0.6858 | 0.6295 | 0.7532 | None |
85
- | 0.5039 | 2.22 | 3000 | 0.5390 | 0.6865 | 0.6167 | 0.7741 | None |
86
- | 0.5038 | 2.59 | 3500 | 0.5231 | 0.6877 | 0.6460 | 0.7353 | None |
87
- | 0.5019 | 2.96 | 4000 | 0.5340 | 0.6905 | 0.6200 | 0.7791 | None |
88
- | 0.481 | 3.32 | 4500 | 0.5323 | 0.6926 | 0.6327 | 0.7650 | None |
89
- | 0.4816 | 3.69 | 5000 | 0.5323 | 0.6932 | 0.6370 | 0.7602 | None |
90
 
91
 
92
  ### Framework versions
 
23
  metrics:
24
  - name: F1
25
  type: f1
26
+ value: 0.6929572707312953
27
  - name: Precision
28
  type: precision
29
+ value: 0.6382018976512677
30
  - name: Recall
31
  type: recall
32
+ value: 0.7579900240162571
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: 0.5309
43
+ - F1: 0.6930
44
+ - Precision: 0.6382
45
+ - Recall: 0.7580
46
  - Support: None
47
 
48
  ## Model description
 
69
  - gradient_accumulation_steps: 2
70
  - total_train_batch_size: 64
71
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
72
+ - lr_scheduler_type: linear
73
  - num_epochs: 30
74
  - label_smoothing_factor: 0.1
75
 
 
77
 
78
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
79
  |:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:|:-------:|
80
+ | 0.5819 | 0.37 | 500 | 0.6020 | 0.6493 | 0.5311 | 0.8352 | None |
81
+ | 0.5611 | 0.74 | 1000 | 0.5814 | 0.6648 | 0.5561 | 0.8263 | None |
82
+ | 0.5467 | 1.11 | 1500 | 0.5570 | 0.6744 | 0.5878 | 0.7911 | None |
83
+ | 0.5345 | 1.48 | 2000 | 0.5228 | 0.6756 | 0.6470 | 0.7068 | None |
84
+ | 0.5299 | 1.85 | 2500 | 0.5283 | 0.6853 | 0.6291 | 0.7526 | None |
85
+ | 0.5054 | 2.22 | 3000 | 0.5395 | 0.6850 | 0.6159 | 0.7717 | None |
86
+ | 0.5052 | 2.59 | 3500 | 0.5236 | 0.6876 | 0.6450 | 0.7362 | None |
87
+ | 0.5033 | 2.96 | 4000 | 0.5338 | 0.6899 | 0.6202 | 0.7774 | None |
88
+ | 0.4839 | 3.32 | 4500 | 0.5339 | 0.6922 | 0.6292 | 0.7691 | None |
89
+ | 0.4846 | 3.69 | 5000 | 0.5309 | 0.6930 | 0.6382 | 0.7580 | None |
90
 
91
 
92
  ### Framework versions
config.toml CHANGED
@@ -1,5 +1,5 @@
1
  [experiment]
2
- name = "binary-31"
3
  type = "binary"
4
 
5
 
@@ -18,7 +18,7 @@ revision = "2caf9ebc5b275737c95f8bb16953288107a7131c"
18
  [training]
19
  num_train_epochs = 30
20
  warmup_steps = 0
21
- lr_scheduler_type = "constant"
22
  learning_rate = 1e-5
23
  per_device_train_batch_size = 32
24
  per_device_eval_batch_size = 32
 
1
  [experiment]
2
+ name = "binary-32"
3
  type = "binary"
4
 
5
 
 
18
  [training]
19
  num_train_epochs = 30
20
  warmup_steps = 0
21
+ lr_scheduler_type = "linear"
22
  learning_rate = 1e-5
23
  per_device_train_batch_size = 32
24
  per_device_eval_batch_size = 32
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b3db586068ea265a8e1c0940d29e45e2d6c7c3fb1d6f36d8ac0934aeba8a437
3
  size 313992076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffa9b265fc5a19baa463bee60e9074ec131fe038226868e561b81b1be06a3607
3
  size 313992076
tokenizer.json CHANGED
@@ -6,7 +6,14 @@
6
  "strategy": "LongestFirst",
7
  "stride": 0
8
  },
9
- "padding": null,
 
 
 
 
 
 
 
10
  "added_tokens": [
11
  {
12
  "id": 0,
 
6
  "strategy": "LongestFirst",
7
  "stride": 0
8
  },
9
+ "padding": {
10
+ "strategy": "BatchLongest",
11
+ "direction": "Right",
12
+ "pad_to_multiple_of": null,
13
+ "pad_id": 0,
14
+ "pad_type_id": 0,
15
+ "pad_token": "[PAD]"
16
+ },
17
  "added_tokens": [
18
  {
19
  "id": 0,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bab686dc26ca97df0bc8fd3e521527ef779b55d1b3a485e5f4b76b3cb1cb4d41
3
  size 4091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc6a77945690f1f63f16259eb40d607f54a53c733180aa51a00d74f6362b8f6f
3
  size 4091