Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -14,7 +14,7 @@ widget:
|
|
14 |
|
15 |
## Introduction
|
16 |
|
17 |
-
This model is a model that was fine-tuned from [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/
|
18 |
|
19 |
## Dataset
|
20 |
|
@@ -33,19 +33,19 @@ FT |2|Geographical feature
|
|
33 |
|
34 |
## Experiment parameter
|
35 |
|
36 |
-
* Pretrained-model : [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/
|
37 |
* Dataset : noisy (Pero OCR)
|
38 |
* Tagging format : IOB2
|
39 |
* Recognised entities : level 2
|
40 |
|
41 |
## Load model from the Hugging Face
|
42 |
|
43 |
-
**Warning** : this model only recognises level-2 entities of dataset. It has to be used with [m1_ind_layers_ocr_cmbert_iob2_level_1](https://huggingface.co/nlpso/m1_ind_layers_ocr_cmbert_iob2_level_1) to recognise nested entities level-1.
|
44 |
|
45 |
```python
|
46 |
from transformers import AutoTokenizer, AutoModelForTokenClassification
|
47 |
|
48 |
-
tokenizer = AutoTokenizer.from_pretrained("m1_ind_layers_ocr_cmbert_iob2_level_2")
|
49 |
-
model = AutoModelForTokenClassification.from_pretrained("m1_ind_layers_ocr_cmbert_iob2_level_2")
|
50 |
|
51 |
|
|
|
14 |
|
15 |
## Introduction
|
16 |
|
17 |
+
This model is a model that was fine-tuned from [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/HueyNemud/das22-10-camembert_pretrained) for **nested NER task** on a nested NER Paris trade directories dataset.
|
18 |
|
19 |
## Dataset
|
20 |
|
|
|
33 |
|
34 |
## Experiment parameter
|
35 |
|
36 |
+
* Pretrained-model : [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/HueyNemud/das22-10-camembert_pretrained)
|
37 |
* Dataset : noisy (Pero OCR)
|
38 |
* Tagging format : IOB2
|
39 |
* Recognised entities : level 2
|
40 |
|
41 |
## Load model from the Hugging Face
|
42 |
|
43 |
+
**Warning 1 ** : this model only recognises level-2 entities of dataset. It has to be used with [m1_ind_layers_ocr_cmbert_iob2_level_1](https://huggingface.co/nlpso/m1_ind_layers_ocr_cmbert_iob2_level_1) to recognise nested entities level-1.
|
44 |
|
45 |
```python
|
46 |
from transformers import AutoTokenizer, AutoModelForTokenClassification
|
47 |
|
48 |
+
tokenizer = AutoTokenizer.from_pretrained(nlpso/"m1_ind_layers_ocr_cmbert_iob2_level_2")
|
49 |
+
model = AutoModelForTokenClassification.from_pretrained(nlpso/"m1_ind_layers_ocr_cmbert_iob2_level_2")
|
50 |
|
51 |
|