Hugging Face's logo

haryoaw
/
scenario-kd-from-scratch-gold-silver-data-hate_speech_filipino-model-xlm-roberta

Transformers
PyTorch
xlm-roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
haryoaw commited on
Commit
b759f14
•
1 Parent(s): e41f1a2

Initial Commit

Browse files
Files changed (4) hide show
  1. README.md +100 -0
  2. config.json +29 -0
  3. pytorch_model.bin +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: xlm-roberta-base
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - hate_speech_filipino
8
+ metrics:
9
+ - accuracy
10
+ - f1
11
+ model-index:
12
+ - name: scenario-kd-from-scratch-gold-silver-data-hate_speech_filipino-model-xlm-roberta
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # scenario-kd-from-scratch-gold-silver-data-hate_speech_filipino-model-xlm-roberta
20
+
21
+ This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the hate_speech_filipino dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.7729
24
+ - Accuracy: 0.7559
25
+ - F1: 0.7334
26
+
27
+ ## Model description
28
+
29
+ More information needed
30
+
31
+ ## Intended uses & limitations
32
+
33
+ More information needed
34
+
35
+ ## Training and evaluation data
36
+
37
+ More information needed
38
+
39
+ ## Training procedure
40
+
41
+ ### Training hyperparameters
42
+
43
+ The following hyperparameters were used during training:
44
+ - learning_rate: 5e-05
45
+ - train_batch_size: 32
46
+ - eval_batch_size: 32
47
+ - seed: 42
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: linear
50
+ - num_epochs: 6969
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
55
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
56
+ | No log | 0.32 | 100 | 3.1640 | 0.6628 | 0.6705 |
57
+ | No log | 0.64 | 200 | 2.7957 | 0.7136 | 0.6553 |
58
+ | No log | 0.96 | 300 | 2.4898 | 0.7141 | 0.7235 |
59
+ | No log | 1.28 | 400 | 2.3972 | 0.7353 | 0.6774 |
60
+ | 3.1684 | 1.6 | 500 | 2.1905 | 0.7476 | 0.7243 |
61
+ | 3.1684 | 1.92 | 600 | 2.1041 | 0.7481 | 0.7406 |
62
+ | 3.1684 | 2.24 | 700 | 2.1459 | 0.7481 | 0.7084 |
63
+ | 3.1684 | 2.56 | 800 | 2.3199 | 0.7457 | 0.6848 |
64
+ | 3.1684 | 2.88 | 900 | 2.0306 | 0.7422 | 0.7461 |
65
+ | 2.009 | 3.19 | 1000 | 2.0210 | 0.7623 | 0.7414 |
66
+ | 2.009 | 3.51 | 1100 | 2.1153 | 0.7554 | 0.7045 |
67
+ | 2.009 | 3.83 | 1200 | 1.9246 | 0.7583 | 0.7452 |
68
+ | 2.009 | 4.15 | 1300 | 1.9316 | 0.7611 | 0.7300 |
69
+ | 2.009 | 4.47 | 1400 | 2.5109 | 0.7547 | 0.7107 |
70
+ | 1.5147 | 4.79 | 1500 | 2.1018 | 0.7339 | 0.7465 |
71
+ | 1.5147 | 5.11 | 1600 | 2.5402 | 0.7344 | 0.6409 |
72
+ | 1.5147 | 5.43 | 1700 | 1.9100 | 0.7602 | 0.7427 |
73
+ | 1.5147 | 5.75 | 1800 | 2.0015 | 0.7519 | 0.7469 |
74
+ | 1.5147 | 6.07 | 1900 | 1.9725 | 0.7389 | 0.7376 |
75
+ | 1.2233 | 6.39 | 2000 | 1.8734 | 0.7545 | 0.7499 |
76
+ | 1.2233 | 6.71 | 2100 | 1.7899 | 0.7677 | 0.7478 |
77
+ | 1.2233 | 7.03 | 2200 | 1.9712 | 0.7583 | 0.7507 |
78
+ | 1.2233 | 7.35 | 2300 | 2.0334 | 0.7545 | 0.7049 |
79
+ | 1.2233 | 7.67 | 2400 | 1.7572 | 0.7583 | 0.7439 |
80
+ | 1.0785 | 7.99 | 2500 | 1.9824 | 0.7455 | 0.7361 |
81
+ | 1.0785 | 8.31 | 2600 | 2.1105 | 0.7306 | 0.7505 |
82
+ | 1.0785 | 8.63 | 2700 | 1.8427 | 0.7521 | 0.7451 |
83
+ | 1.0785 | 8.95 | 2800 | 1.9739 | 0.7569 | 0.7141 |
84
+ | 1.0785 | 9.27 | 2900 | 1.8361 | 0.7557 | 0.7401 |
85
+ | 0.9726 | 9.58 | 3000 | 1.8516 | 0.7564 | 0.7216 |
86
+ | 0.9726 | 9.9 | 3100 | 2.0626 | 0.7346 | 0.7495 |
87
+ | 0.9726 | 10.22 | 3200 | 1.8399 | 0.7651 | 0.7407 |
88
+ | 0.9726 | 10.54 | 3300 | 1.9179 | 0.7609 | 0.7201 |
89
+ | 0.9726 | 10.86 | 3400 | 1.7367 | 0.7625 | 0.7328 |
90
+ | 0.9237 | 11.18 | 3500 | 2.0266 | 0.7342 | 0.7462 |
91
+ | 0.9237 | 11.5 | 3600 | 2.3553 | 0.7094 | 0.7371 |
92
+ | 0.9237 | 11.82 | 3700 | 1.7729 | 0.7559 | 0.7334 |
93
+
94
+
95
+ ### Framework versions
96
+
97
+ - Transformers 4.33.3
98
+ - Pytorch 2.0.1
99
+ - Datasets 2.14.5
100
+ - Tokenizers 0.13.3
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "xlm-roberta-base",
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassificationKD"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "xlm-roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 6,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "problem_type": "single_label_classification",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.33.3",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 250002
29
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:771ac2628a0ecb61d4423076f8b5ecef7bb382d2d59b20f12c6c13910378b4db
3
+ size 942107569
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2683b09eccc308b369ec83cdd17c3f08e6a6c209a0d7e191b0a819b89fa9f1a9
3
+ size 4219