Hugging Face's logo

Alfanatasya
/
results_indobert-large-p2_preprocessing_tuning

Text Classification
Transformers
Safetensors
bert
Generated from Trainer
Model card Files
xet
Community
Alfanatasya commited on
Commit
c314f27
·
verified ·
1 Parent(s): c3bd11a

indobert-large-p2_preprocessing_tuning

Browse files
Files changed (7) hide show
  1. README.md +14 -17
  2. config.json +2 -2
  3. final_model/config.json +2 -2
  4. final_model/model.safetensors +1 -1
  5. final_model/training_args.bin +1 -1
  6. model.safetensors +1 -1
  7. training_args.bin +1 -1
README.md CHANGED
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [indobenchmark/indobert-large-p2](https://huggingface.co/indobenchmark/indobert-large-p2) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 1.2080
25
- - Accuracy: 0.7727
26
- - Precision: 0.7824
27
- - Recall: 0.7786
28
- - F1: 0.7792
29
 
30
  ## Model description
31
 
@@ -44,12 +44,11 @@ More information needed
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
- - learning_rate: 4.574921995691458e-05
48
  - train_batch_size: 32
49
  - eval_batch_size: 32
50
  - seed: 42
51
- - optimizer: Use OptimizerNames.ADAFACTOR and the args are:
52
- No additional optimizer arguments
53
  - lr_scheduler_type: linear
54
  - num_epochs: 20
55
 
@@ -57,15 +56,13 @@ No additional optimizer arguments
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
59
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
60
- | 0.8626 | 1.0 | 111 | 0.7792 | 0.7455 | 0.7639 | 0.7500 | 0.7359 |
61
- | 0.3943 | 2.0 | 222 | 0.8798 | 0.7364 | 0.7480 | 0.7483 | 0.7394 |
62
- | 0.1861 | 3.0 | 333 | 1.0242 | 0.7636 | 0.7808 | 0.7597 | 0.7672 |
63
- | 0.0715 | 4.0 | 444 | 1.2080 | 0.7727 | 0.7824 | 0.7786 | 0.7792 |
64
- | 0.0283 | 5.0 | 555 | 1.4991 | 0.7591 | 0.7709 | 0.7658 | 0.7638 |
65
- | 0.0248 | 6.0 | 666 | 1.5037 | 0.7591 | 0.7721 | 0.7576 | 0.7612 |
66
- | 0.0071 | 7.0 | 777 | 1.6257 | 0.7659 | 0.7676 | 0.7760 | 0.7696 |
67
- | 0.0024 | 8.0 | 888 | 1.7610 | 0.7614 | 0.7609 | 0.7723 | 0.7644 |
68
- | 0.0022 | 9.0 | 999 | 1.8253 | 0.7568 | 0.7563 | 0.7687 | 0.7597 |
69
 
70
 
71
  ### Framework versions
 
21
 
22
  This model is a fine-tuned version of [indobenchmark/indobert-large-p2](https://huggingface.co/indobenchmark/indobert-large-p2) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.6673
25
+ - Accuracy: 0.7841
26
+ - Precision: 0.7920
27
+ - Recall: 0.7918
28
+ - F1: 0.7901
29
 
30
  ## Model description
31
 
 
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
+ - learning_rate: 2.3352320097915953e-05
48
  - train_batch_size: 32
49
  - eval_batch_size: 32
50
  - seed: 42
51
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 
52
  - lr_scheduler_type: linear
53
  - num_epochs: 20
54
 
 
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
58
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
59
+ | 1.2207 | 1.0 | 111 | 0.7383 | 0.7409 | 0.7463 | 0.7491 | 0.7435 |
60
+ | 0.6702 | 2.0 | 222 | 0.6673 | 0.7841 | 0.7920 | 0.7918 | 0.7901 |
61
+ | 0.4953 | 3.0 | 333 | 0.7161 | 0.7636 | 0.7707 | 0.7722 | 0.7711 |
62
+ | 0.3754 | 4.0 | 444 | 0.8318 | 0.75 | 0.7552 | 0.7657 | 0.7569 |
63
+ | 0.2769 | 5.0 | 555 | 0.8916 | 0.7591 | 0.7587 | 0.7732 | 0.7642 |
64
+ | 0.2039 | 6.0 | 666 | 0.9693 | 0.7432 | 0.7533 | 0.7589 | 0.7524 |
65
+ | 0.1525 | 7.0 | 777 | 1.0838 | 0.7477 | 0.7431 | 0.7610 | 0.7471 |
 
 
66
 
67
 
68
  ### Framework versions
config.json CHANGED
@@ -3,11 +3,11 @@
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
6
- "attention_probs_dropout_prob": 0.1,
7
  "classifier_dropout": null,
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
- "hidden_dropout_prob": 0.1,
11
  "hidden_size": 1024,
12
  "id2label": {
13
  "0": "LABEL_0",
 
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
6
+ "attention_probs_dropout_prob": 0.3,
7
  "classifier_dropout": null,
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.3,
11
  "hidden_size": 1024,
12
  "id2label": {
13
  "0": "LABEL_0",
final_model/config.json CHANGED
@@ -3,11 +3,11 @@
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
6
- "attention_probs_dropout_prob": 0.1,
7
  "classifier_dropout": null,
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
- "hidden_dropout_prob": 0.1,
11
  "hidden_size": 1024,
12
  "id2label": {
13
  "0": "LABEL_0",
 
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
6
+ "attention_probs_dropout_prob": 0.3,
7
  "classifier_dropout": null,
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.3,
11
  "hidden_size": 1024,
12
  "id2label": {
13
  "0": "LABEL_0",
final_model/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ec3179bcc7bf1ad095411bdb9e8dfc46ad1e885175627d9e9ab1d8b753795fc
3
  size 1340635060
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12dce05ada3ba7f7a9b691e2a5449f92abdba7faedc68fbdff9a4b789cf0775c
3
  size 1340635060
final_model/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4c6f1017d41aac00d808abc885e3baf9c6ea1fdda0db35908bdd2a827ba3b10
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564cc7aa93f14e1bb96404a32c2266f2447c3039a96fa866935213511d4991a0
3
  size 5304
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ec3179bcc7bf1ad095411bdb9e8dfc46ad1e885175627d9e9ab1d8b753795fc
3
  size 1340635060
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12dce05ada3ba7f7a9b691e2a5449f92abdba7faedc68fbdff9a4b789cf0775c
3
  size 1340635060
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4c6f1017d41aac00d808abc885e3baf9c6ea1fdda0db35908bdd2a827ba3b10
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564cc7aa93f14e1bb96404a32c2266f2447c3039a96fa866935213511d4991a0
3
  size 5304