End of training
Browse files- README.md +107 -0
- model.safetensors +1 -1
README.md
ADDED
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: google-bert/bert-base-multilingual-cased
|
5 |
+
tags:
|
6 |
+
- named-entity-recognition
|
7 |
+
- lumasaba
|
8 |
+
- african-language
|
9 |
+
- pii-detection
|
10 |
+
- token-classification
|
11 |
+
- generated_from_trainer
|
12 |
+
datasets:
|
13 |
+
- Beijuka/Multilingual_PII_NER_dataset
|
14 |
+
metrics:
|
15 |
+
- precision
|
16 |
+
- recall
|
17 |
+
- f1
|
18 |
+
- accuracy
|
19 |
+
model-index:
|
20 |
+
- name: multilingual-google-bert/bert-base-multilingual-cased-lumasaba-ner-v1
|
21 |
+
results:
|
22 |
+
- task:
|
23 |
+
name: Token Classification
|
24 |
+
type: token-classification
|
25 |
+
dataset:
|
26 |
+
name: Beijuka/Multilingual_PII_NER_dataset
|
27 |
+
type: Beijuka/Multilingual_PII_NER_dataset
|
28 |
+
args: 'split: train+validation+test'
|
29 |
+
metrics:
|
30 |
+
- name: Precision
|
31 |
+
type: precision
|
32 |
+
value: 0.9705014749262537
|
33 |
+
- name: Recall
|
34 |
+
type: recall
|
35 |
+
value: 0.9529326574945691
|
36 |
+
- name: F1
|
37 |
+
type: f1
|
38 |
+
value: 0.9616368286445013
|
39 |
+
- name: Accuracy
|
40 |
+
type: accuracy
|
41 |
+
value: 0.9603766182816791
|
42 |
+
---
|
43 |
+
|
44 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
45 |
+
should probably proofread and complete it, then remove this comment. -->
|
46 |
+
|
47 |
+
# multilingual-google-bert/bert-base-multilingual-cased-lumasaba-ner-v1
|
48 |
+
|
49 |
+
This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on the Beijuka/Multilingual_PII_NER_dataset dataset.
|
50 |
+
It achieves the following results on the evaluation set:
|
51 |
+
- Loss: 0.3203
|
52 |
+
- Precision: 0.9705
|
53 |
+
- Recall: 0.9529
|
54 |
+
- F1: 0.9616
|
55 |
+
- Accuracy: 0.9604
|
56 |
+
|
57 |
+
## Model description
|
58 |
+
|
59 |
+
More information needed
|
60 |
+
|
61 |
+
## Intended uses & limitations
|
62 |
+
|
63 |
+
More information needed
|
64 |
+
|
65 |
+
## Training and evaluation data
|
66 |
+
|
67 |
+
More information needed
|
68 |
+
|
69 |
+
## Training procedure
|
70 |
+
|
71 |
+
### Training hyperparameters
|
72 |
+
|
73 |
+
The following hyperparameters were used during training:
|
74 |
+
- learning_rate: 5e-05
|
75 |
+
- train_batch_size: 8
|
76 |
+
- eval_batch_size: 8
|
77 |
+
- seed: 42
|
78 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
79 |
+
- lr_scheduler_type: linear
|
80 |
+
- num_epochs: 20
|
81 |
+
|
82 |
+
### Training results
|
83 |
+
|
84 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
85 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
86 |
+
| No log | 1.0 | 398 | 0.6266 | 0.8401 | 0.8225 | 0.8312 | 0.8062 |
|
87 |
+
| 1.0576 | 2.0 | 796 | 0.3751 | 0.9033 | 0.8891 | 0.8962 | 0.8859 |
|
88 |
+
| 0.3626 | 3.0 | 1194 | 0.3664 | 0.9336 | 0.9273 | 0.9305 | 0.9163 |
|
89 |
+
| 0.1629 | 4.0 | 1592 | 0.4134 | 0.9381 | 0.9303 | 0.9342 | 0.9244 |
|
90 |
+
| 0.1629 | 5.0 | 1990 | 0.3573 | 0.9497 | 0.9476 | 0.9486 | 0.9417 |
|
91 |
+
| 0.0925 | 6.0 | 2388 | 0.4060 | 0.9501 | 0.9416 | 0.9458 | 0.9434 |
|
92 |
+
| 0.0516 | 7.0 | 2786 | 0.3767 | 0.9371 | 0.9483 | 0.9427 | 0.9377 |
|
93 |
+
| 0.0409 | 8.0 | 3184 | 0.4152 | 0.9450 | 0.9528 | 0.9489 | 0.9409 |
|
94 |
+
| 0.0389 | 9.0 | 3582 | 0.3901 | 0.9624 | 0.9386 | 0.9503 | 0.9458 |
|
95 |
+
| 0.0389 | 10.0 | 3980 | 0.4474 | 0.9388 | 0.9536 | 0.9461 | 0.9426 |
|
96 |
+
| 0.0212 | 11.0 | 4378 | 0.3165 | 0.9591 | 0.9663 | 0.9627 | 0.9547 |
|
97 |
+
| 0.0167 | 12.0 | 4776 | 0.3941 | 0.9590 | 0.9633 | 0.9611 | 0.9543 |
|
98 |
+
| 0.0199 | 13.0 | 5174 | 0.4243 | 0.9496 | 0.9588 | 0.9542 | 0.9478 |
|
99 |
+
| 0.0156 | 14.0 | 5572 | 0.4842 | 0.9539 | 0.9618 | 0.9579 | 0.9494 |
|
100 |
+
|
101 |
+
|
102 |
+
### Framework versions
|
103 |
+
|
104 |
+
- Transformers 4.55.4
|
105 |
+
- Pytorch 2.8.0+cu126
|
106 |
+
- Datasets 4.0.0
|
107 |
+
- Tokenizers 0.21.4
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 709163956
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8c6f6d060634946df0ea99f8757e8447c54fb030942b3ef864dc6b48507e5f65
|
3 |
size 709163956
|