Model save
Browse files
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-base-3
|
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-base-3
|
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: 0.2918
|
24 |
+
- Precision: 0.7974
|
25 |
+
- Recall: 0.8847
|
26 |
+
- F1: 0.8388
|
27 |
+
- Accuracy: 0.9619
|
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 |
+
| 0.3817 | 1.0 | 106 | 0.1442 | 0.7266 | 0.8732 | 0.7932 | 0.9515 |
|
59 |
+
| 0.1266 | 2.0 | 212 | 0.1385 | 0.7381 | 0.8934 | 0.8083 | 0.9551 |
|
60 |
+
| 0.087 | 3.0 | 318 | 0.1367 | 0.7512 | 0.8790 | 0.8101 | 0.9568 |
|
61 |
+
| 0.0528 | 4.0 | 424 | 0.1468 | 0.7732 | 0.8646 | 0.8163 | 0.9595 |
|
62 |
+
| 0.0424 | 5.0 | 530 | 0.1664 | 0.7899 | 0.8559 | 0.8216 | 0.9607 |
|
63 |
+
| 0.0275 | 6.0 | 636 | 0.2044 | 0.7714 | 0.8847 | 0.8242 | 0.9583 |
|
64 |
+
| 0.019 | 7.0 | 742 | 0.2377 | 0.7410 | 0.8905 | 0.8089 | 0.9554 |
|
65 |
+
| 0.0145 | 8.0 | 848 | 0.2432 | 0.7758 | 0.8876 | 0.8280 | 0.9588 |
|
66 |
+
| 0.0102 | 9.0 | 954 | 0.2287 | 0.8109 | 0.9020 | 0.8540 | 0.9641 |
|
67 |
+
| 0.0067 | 10.0 | 1060 | 0.2430 | 0.8026 | 0.8905 | 0.8443 | 0.9617 |
|
68 |
+
| 0.0064 | 11.0 | 1166 | 0.2675 | 0.7943 | 0.8905 | 0.8397 | 0.9602 |
|
69 |
+
| 0.0046 | 12.0 | 1272 | 0.2743 | 0.7828 | 0.8934 | 0.8345 | 0.9619 |
|
70 |
+
| 0.0034 | 13.0 | 1378 | 0.2666 | 0.7995 | 0.8963 | 0.8451 | 0.9619 |
|
71 |
+
| 0.0036 | 14.0 | 1484 | 0.2606 | 0.8117 | 0.8818 | 0.8453 | 0.9634 |
|
72 |
+
| 0.0027 | 15.0 | 1590 | 0.2862 | 0.7913 | 0.8963 | 0.8405 | 0.9627 |
|
73 |
+
| 0.0016 | 16.0 | 1696 | 0.2793 | 0.8021 | 0.8876 | 0.8427 | 0.9629 |
|
74 |
+
| 0.0012 | 17.0 | 1802 | 0.2951 | 0.7949 | 0.8934 | 0.8412 | 0.9622 |
|
75 |
+
| 0.0012 | 18.0 | 1908 | 0.2930 | 0.7938 | 0.8876 | 0.8381 | 0.9617 |
|
76 |
+
| 0.0014 | 19.0 | 2014 | 0.2953 | 0.7912 | 0.8847 | 0.8354 | 0.9612 |
|
77 |
+
| 0.0007 | 20.0 | 2120 | 0.2918 | 0.7974 | 0.8847 | 0.8388 | 0.9619 |
|
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
|