Commit
•
3c7ba90
1
Parent(s):
b7faf97
End of training
Browse files
README.md
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: indolem/indobertweet-base-uncased
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
model-index:
|
7 |
+
- name: classification-hate-speech-DE-14-twit
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# classification-hate-speech-DE-14-twit
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [indolem/indobertweet-base-uncased](https://huggingface.co/indolem/indobertweet-base-uncased) on the None dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 2.7131
|
19 |
+
- F1 macro: 0.4211
|
20 |
+
- Weighted: 0.6318
|
21 |
+
- Balanced accuracy: 0.5466
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 5e-05
|
41 |
+
- train_batch_size: 8
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- num_epochs: 14
|
47 |
+
|
48 |
+
### Training results
|
49 |
+
|
50 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 macro | Weighted | Balanced accuracy |
|
51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------------:|
|
52 |
+
| 1.2526 | 1.0 | 152 | 1.0691 | 0.4123 | 0.6723 | 0.4918 |
|
53 |
+
| 0.8027 | 2.0 | 304 | 1.0656 | 0.4134 | 0.6745 | 0.4853 |
|
54 |
+
| 0.2427 | 3.0 | 456 | 1.6157 | 0.4341 | 0.6215 | 0.5589 |
|
55 |
+
| 0.0699 | 4.0 | 608 | 1.8798 | 0.3919 | 0.6305 | 0.5128 |
|
56 |
+
| 0.0721 | 5.0 | 760 | 2.5604 | 0.3917 | 0.5534 | 0.5298 |
|
57 |
+
| 0.0514 | 6.0 | 912 | 2.4451 | 0.3949 | 0.6116 | 0.5375 |
|
58 |
+
| 0.002 | 7.0 | 1064 | 2.6198 | 0.4128 | 0.6078 | 0.5445 |
|
59 |
+
| 0.0161 | 8.0 | 1216 | 2.6067 | 0.4185 | 0.6181 | 0.5441 |
|
60 |
+
| 0.0007 | 9.0 | 1368 | 2.5439 | 0.4209 | 0.6326 | 0.5436 |
|
61 |
+
| 0.0006 | 10.0 | 1520 | 2.6588 | 0.4199 | 0.6296 | 0.5476 |
|
62 |
+
| 0.0005 | 11.0 | 1672 | 2.6799 | 0.4210 | 0.6321 | 0.5481 |
|
63 |
+
| 0.0004 | 12.0 | 1824 | 2.6578 | 0.4223 | 0.6358 | 0.5458 |
|
64 |
+
| 0.0004 | 13.0 | 1976 | 2.6912 | 0.4237 | 0.6369 | 0.5476 |
|
65 |
+
| 0.0005 | 14.0 | 2128 | 2.7131 | 0.4211 | 0.6318 | 0.5466 |
|
66 |
+
|
67 |
+
|
68 |
+
### Framework versions
|
69 |
+
|
70 |
+
- Transformers 4.41.2
|
71 |
+
- Pytorch 2.3.0+cu121
|
72 |
+
- Datasets 2.20.0
|
73 |
+
- Tokenizers 0.19.1
|
config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "indolem/indobertweet-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_ids": 0,
|
10 |
+
"gradient_checkpointing": false,
|
11 |
+
"hidden_act": "gelu",
|
12 |
+
"hidden_dropout_prob": 0.1,
|
13 |
+
"hidden_size": 768,
|
14 |
+
"id2label": {
|
15 |
+
"0": "LABEL_0",
|
16 |
+
"1": "LABEL_1",
|
17 |
+
"2": "LABEL_2",
|
18 |
+
"3": "LABEL_3",
|
19 |
+
"4": "LABEL_4",
|
20 |
+
"5": "LABEL_5",
|
21 |
+
"6": "LABEL_6",
|
22 |
+
"7": "LABEL_7"
|
23 |
+
},
|
24 |
+
"initializer_range": 0.02,
|
25 |
+
"intermediate_size": 3072,
|
26 |
+
"label2id": {
|
27 |
+
"LABEL_0": 0,
|
28 |
+
"LABEL_1": 1,
|
29 |
+
"LABEL_2": 2,
|
30 |
+
"LABEL_3": 3,
|
31 |
+
"LABEL_4": 4,
|
32 |
+
"LABEL_5": 5,
|
33 |
+
"LABEL_6": 6,
|
34 |
+
"LABEL_7": 7
|
35 |
+
},
|
36 |
+
"layer_norm_eps": 1e-12,
|
37 |
+
"max_position_embeddings": 512,
|
38 |
+
"model_type": "bert",
|
39 |
+
"num_attention_heads": 12,
|
40 |
+
"num_hidden_layers": 12,
|
41 |
+
"output_past": true,
|
42 |
+
"pad_token_id": 0,
|
43 |
+
"position_embedding_type": "absolute",
|
44 |
+
"problem_type": "single_label_classification",
|
45 |
+
"torch_dtype": "float32",
|
46 |
+
"transformers_version": "4.41.2",
|
47 |
+
"type_vocab_size": 2,
|
48 |
+
"use_cache": true,
|
49 |
+
"vocab_size": 31923
|
50 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:545e833a374547d3619abd2dfe259c513b5d90e5d79458d7a263aebd059e2962
|
3 |
+
size 442280976
|
runs/Jun19_15-55-03_6e71a2621ca0/events.out.tfevents.1718812504.6e71a2621ca0.1065.12
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d751dee6b571d333cfe803d91bf11724111af3a1353cffe68aa9ee42b8631889
|
3 |
+
size 30160
|
runs/Jun19_15-57-07_6e71a2621ca0/events.out.tfevents.1718812628.6e71a2621ca0.1065.13
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f996cda280b03aad4f02f6c7948a40ab9e9006a101ab6ba60ad170477adac70e
|
3 |
+
size 56488
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:705e6429aea9300ec1256c130e9965937d5c8089b97050262595d17023989fdd
|
3 |
+
size 5176
|