Hugging Face's logo

apwic
/
nerugm-lora-r16-2

TensorBoard
Indonesian
Generated from Trainer
Model card Files Metrics Community
apwic commited on
Commit
a3a701f
•
1 Parent(s): 3f529b3

Model save

Browse files
Files changed (6) hide show
  1. README.md +85 -0
  2. nerugm-lora/adapter_config.json +26 -0
  3. nerugm-lora/head_config.json +23 -0
  4. nerugm-lora/pytorch_adapter.bin +3 -0
  5. nerugm-lora/pytorch_model_head.bin +3 -0
  6. runs/May27_05-21-21_indolem-petl-vm/events.out.tfevents.1716787288.indolem-petl-vm.3332526.0 +2 -2
README.md ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: indolem/indobert-base-uncased
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - precision
8
+ - recall
9
+ - f1
10
+ - accuracy
11
+ model-index:
12
+ - name: nerugm-lora-r16-2
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
+ # nerugm-lora-r16-2
20
+
21
+ This model is a fine-tuned version of [indolem/indobert-base-uncased](https://huggingface.co/indolem/indobert-base-uncased) on an unknown dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.1626
24
+ - Precision: 0.6848
25
+ - Recall: 0.8525
26
+ - F1: 0.7595
27
+ - Accuracy: 0.9472
28
+
29
+ ## Model description
30
+
31
+ More information needed
32
+
33
+ ## Intended uses & limitations
34
+
35
+ More information needed
36
+
37
+ ## Training and evaluation data
38
+
39
+ More information needed
40
+
41
+ ## Training procedure
42
+
43
+ ### Training hyperparameters
44
+
45
+ The following hyperparameters were used during training:
46
+ - learning_rate: 5e-05
47
+ - train_batch_size: 16
48
+ - eval_batch_size: 64
49
+ - seed: 42
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: linear
52
+ - num_epochs: 20.0
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
57
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
58
+ | 1.1483 | 1.0 | 106 | 0.6900 | 0.0 | 0.0 | 0.0 | 0.8449 |
59
+ | 0.6875 | 2.0 | 212 | 0.5737 | 0.0 | 0.0 | 0.0 | 0.8464 |
60
+ | 0.5874 | 3.0 | 318 | 0.4661 | 0.2692 | 0.0619 | 0.1007 | 0.8634 |
61
+ | 0.4729 | 4.0 | 424 | 0.3599 | 0.4753 | 0.3127 | 0.3772 | 0.8982 |
62
+ | 0.3692 | 5.0 | 530 | 0.2940 | 0.5714 | 0.6136 | 0.5917 | 0.9247 |
63
+ | 0.3058 | 6.0 | 636 | 0.2527 | 0.6110 | 0.7227 | 0.6622 | 0.9335 |
64
+ | 0.2636 | 7.0 | 742 | 0.2246 | 0.6402 | 0.7611 | 0.6954 | 0.9375 |
65
+ | 0.24 | 8.0 | 848 | 0.2091 | 0.6578 | 0.8053 | 0.7241 | 0.9417 |
66
+ | 0.2228 | 9.0 | 954 | 0.1986 | 0.6404 | 0.8142 | 0.7169 | 0.9402 |
67
+ | 0.2105 | 10.0 | 1060 | 0.1821 | 0.6611 | 0.8230 | 0.7332 | 0.9417 |
68
+ | 0.2007 | 11.0 | 1166 | 0.1794 | 0.6675 | 0.8289 | 0.7395 | 0.9432 |
69
+ | 0.195 | 12.0 | 1272 | 0.1808 | 0.6597 | 0.8407 | 0.7393 | 0.9430 |
70
+ | 0.19 | 13.0 | 1378 | 0.1690 | 0.6787 | 0.8289 | 0.7463 | 0.9460 |
71
+ | 0.1835 | 14.0 | 1484 | 0.1631 | 0.6870 | 0.8289 | 0.7513 | 0.9477 |
72
+ | 0.1821 | 15.0 | 1590 | 0.1671 | 0.6835 | 0.8407 | 0.7540 | 0.9472 |
73
+ | 0.1774 | 16.0 | 1696 | 0.1668 | 0.6896 | 0.8584 | 0.7648 | 0.9472 |
74
+ | 0.1764 | 17.0 | 1802 | 0.1635 | 0.6899 | 0.8466 | 0.7603 | 0.9477 |
75
+ | 0.1729 | 18.0 | 1908 | 0.1654 | 0.6856 | 0.8555 | 0.7612 | 0.9472 |
76
+ | 0.1726 | 19.0 | 2014 | 0.1628 | 0.6872 | 0.8555 | 0.7622 | 0.9477 |
77
+ | 0.1684 | 20.0 | 2120 | 0.1626 | 0.6848 | 0.8525 | 0.7595 | 0.9472 |
78
+
79
+
80
+ ### Framework versions
81
+
82
+ - Transformers 4.39.3
83
+ - Pytorch 2.3.0+cu121
84
+ - Datasets 2.19.1
85
+ - Tokenizers 0.15.2
nerugm-lora/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "alpha": 8,
4
+ "architecture": "lora",
5
+ "attn_matrices": [
6
+ "q",
7
+ "v"
8
+ ],
9
+ "composition_mode": "add",
10
+ "dropout": 0.0,
11
+ "init_weights": "lora",
12
+ "intermediate_lora": false,
13
+ "leave_out": [],
14
+ "output_lora": false,
15
+ "r": 16,
16
+ "selfattn_lora": true,
17
+ "use_gating": false
18
+ },
19
+ "config_id": "141b248112091265",
20
+ "hidden_size": 768,
21
+ "model_class": "BertForTokenClassification",
22
+ "model_name": "indolem/indobert-base-uncased",
23
+ "model_type": "bert",
24
+ "name": "nerugm-lora",
25
+ "version": "0.2.0"
26
+ }
nerugm-lora/head_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": null,
3
+ "hidden_size": 768,
4
+ "label2id": {
5
+ "B-LOCATION": 0,
6
+ "B-ORGANIZATION": 1,
7
+ "B-PERSON": 2,
8
+ "B-QUANTITY": 3,
9
+ "B-TIME": 4,
10
+ "I-LOCATION": 5,
11
+ "I-ORGANIZATION": 6,
12
+ "I-PERSON": 7,
13
+ "I-QUANTITY": 8,
14
+ "I-TIME": 9,
15
+ "O": 10
16
+ },
17
+ "model_class": "BertForTokenClassification",
18
+ "model_name": "indolem/indobert-base-uncased",
19
+ "model_type": "bert",
20
+ "name": null,
21
+ "num_labels": 11,
22
+ "version": "0.2.0"
23
+ }
nerugm-lora/pytorch_adapter.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d64a38c000a3e57d01ac55eaf4f72de600a13bdab90ed6aa2ccecb283c835473
3
+ size 2376998
nerugm-lora/pytorch_model_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14aed6c95b3282caeab0114ee62daae14d7ec7324bb060a2c0144186b7bb0a38
3
+ size 35354
runs/May27_05-21-21_indolem-petl-vm/events.out.tfevents.1716787288.indolem-petl-vm.3332526.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:31f9685164a4bc49b04c836452e36d87ed1a3bdabc8cebeb1f6f912f087a70db
3
- size 18177
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cf1b07b8fd1bd549afc0c4832b483541f531e84a158a70d26adff9bc0ad39d8
3
+ size 19214