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
2fd6b3a
·
verified ·
1 Parent(s): ee0cc92

binary-28

Browse files
Files changed (4) hide show
  1. README.md +23 -17
  2. config.toml +3 -3
  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: 0.36679201619901647
27
  - name: Precision
28
  type: precision
29
- value: 0.8447701532311792
30
  - name: Recall
31
  type: recall
32
- value: 0.23425087751708848
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: 0.4531
43
- - F1: 0.3668
44
- - Precision: 0.8448
45
- - Recall: 0.2343
46
  - Support: None
47
 
48
  ## Model description
@@ -78,16 +78,22 @@ The following hyperparameters were used during training:
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
80
  |:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:|:-------:|
81
- | 0.8428 | 0.06 | 500 | 0.5318 | 0.4246 | 0.6560 | 0.3139 | None |
82
- | 0.6948 | 0.13 | 1000 | 0.4850 | 0.4097 | 0.7445 | 0.2827 | None |
83
- | 0.684 | 0.19 | 1500 | 0.4674 | 0.3861 | 0.7563 | 0.2592 | None |
84
- | 0.5896 | 0.25 | 2000 | 0.4870 | 0.2807 | 0.8451 | 0.1683 | None |
85
- | 0.5585 | 0.31 | 2500 | 0.4591 | 0.5076 | 0.7354 | 0.3876 | None |
86
- | 0.5371 | 0.38 | 3000 | 0.4484 | 0.4326 | 0.7993 | 0.2965 | None |
87
- | 0.5286 | 0.44 | 3500 | 0.4479 | 0.4129 | 0.8212 | 0.2758 | None |
88
- | 0.5071 | 0.5 | 4000 | 0.4433 | 0.4647 | 0.7822 | 0.3305 | None |
89
- | 0.5043 | 0.56 | 4500 | 0.4799 | 0.2584 | 0.8539 | 0.1522 | None |
90
- | 0.5149 | 0.63 | 5000 | 0.4531 | 0.3668 | 0.8448 | 0.2343 | None |
 
 
 
 
 
 
91
 
92
 
93
  ### Framework versions
 
23
  metrics:
24
  - name: F1
25
  type: f1
26
+ value: 0.6658038273088657
27
  - name: Precision
28
  type: precision
29
+ value: 0.6663582531458179
30
  - name: Recall
31
  type: recall
32
+ value: 0.6652503232957695
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: 0.6658
44
+ - Precision: 0.6664
45
+ - Recall: 0.6653
46
  - Support: None
47
 
48
  ## Model description
 
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Support |
80
  |:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:|:-------:|
81
+ | 1.0669 | 0.09 | 500 | 1.0365 | 0.5699 | 0.4064 | 0.9531 | None |
82
+ | 0.8978 | 0.18 | 1000 | 1.1794 | 0.5709 | 0.4036 | 0.9756 | None |
83
+ | 0.8142 | 0.28 | 1500 | 0.5742 | 0.6380 | 0.5748 | 0.7168 | None |
84
+ | 0.7269 | 0.37 | 2000 | 0.5638 | 0.6455 | 0.6096 | 0.6858 | None |
85
+ | 0.7215 | 0.46 | 2500 | 0.5475 | 0.6454 | 0.6287 | 0.6630 | None |
86
+ | 0.6828 | 0.55 | 3000 | 0.5702 | 0.6528 | 0.5803 | 0.7460 | None |
87
+ | 0.6857 | 0.65 | 3500 | 0.5348 | 0.6081 | 0.6636 | 0.5612 | None |
88
+ | 0.6381 | 0.74 | 4000 | 0.5609 | 0.6654 | 0.5913 | 0.7608 | None |
89
+ | 0.6476 | 0.83 | 4500 | 0.5565 | 0.6662 | 0.5972 | 0.7532 | None |
90
+ | 0.6124 | 0.92 | 5000 | 0.5468 | 0.6662 | 0.6114 | 0.7318 | None |
91
+ | 0.6292 | 1.02 | 5500 | 0.5409 | 0.6737 | 0.6217 | 0.7353 | None |
92
+ | 0.5904 | 1.11 | 6000 | 0.5565 | 0.6736 | 0.6036 | 0.7621 | None |
93
+ | 0.5926 | 1.2 | 6500 | 0.6091 | 0.6668 | 0.5439 | 0.8614 | None |
94
+ | 0.5962 | 1.29 | 7000 | 0.6201 | 0.6650 | 0.5368 | 0.8736 | None |
95
+ | 0.6056 | 1.39 | 7500 | 0.5477 | 0.6775 | 0.5975 | 0.7824 | None |
96
+ | 0.582 | 1.48 | 8000 | 0.5203 | 0.6658 | 0.6664 | 0.6653 | None |
97
 
98
 
99
  ### Framework versions
config.toml CHANGED
@@ -1,5 +1,5 @@
1
  [experiment]
2
- name = "binary-27"
3
  type = "binary"
4
 
5
 
@@ -7,7 +7,7 @@ type = "binary"
7
  path = "thejosango/nuha-dataset"
8
  dataset_revision = "main"
9
  augment_ratio = 0.0
10
- undersampling_strategy = false
11
 
12
 
13
  [model]
@@ -25,6 +25,6 @@ per_device_eval_batch_size = 8
25
  gradient_accumulation_steps = 2
26
  weight_decay = 0.01
27
  label_smoothing_factor = 0.1
28
- weighted_loss = true
29
  early_stopping_patience = 5
30
  early_stopping_threshold = 0.005
 
1
  [experiment]
2
+ name = "binary-28"
3
  type = "binary"
4
 
5
 
 
7
  path = "thejosango/nuha-dataset"
8
  dataset_revision = "main"
9
  augment_ratio = 0.0
10
+ undersampling_strategy = "majority"
11
 
12
 
13
  [model]
 
25
  gradient_accumulation_steps = 2
26
  weight_decay = 0.01
27
  label_smoothing_factor = 0.1
28
+ weighted_loss = false
29
  early_stopping_patience = 5
30
  early_stopping_threshold = 0.005
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:587e8391854237b42317246fd0b723aee9b71b6d155d6fa2f5f78109242cf9f9
3
  size 313992076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f5f7647297f018fb5effbd25ca4261bfa048f9229c4a3b76e8b39bd0b07841e
3
  size 313992076
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12ec62d67b80d950a507d49d7653460381b8a047dc923e1e97dc8a60d3a0d858
3
  size 4091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f9eac55e3483cca826ec9ff4f42c7edc38f11182cb9243e70a887c7e5c0598c
3
  size 4091