ner-data-processor 0.2

A NER Data Preparing Tool

NER_DATA_PREPROCESSING :

Collecting Data For NER:

this tool helps to create a ner and corefer dataset easily . To train a Token classification and corefer resolution need a dataset. it not like a raw dataset. we want to convert text (sentence) to required format. lets see how this framework/library used in your project. lets go ...

Step - 1:

1. install via git

Download :

 ```bash
 git clone https://github.com/rajboopathiking/NER_DATA_PREPROCESSING.git
 ```

optional (if you already in correct folder)

 ```bash 
 cd NER_DATA_PREPROCESSING
 ```

requirements.txt -->> installation :

```bash
pip install requirements.txt
```

2 ) install via pypi

```bash
pip install ner-data-processor
```
```python
from ner-data-processor.Ner_Data_Preparation import Custom_Ner_Dataset
ner = Custom_Ner_Dataset()
```

Step - 2:

DataSet Format : pandas Dataframe with text(Arun Kumar Jagatramka vs Ultrabulk AS ) and exact word and entity (Arun Kumar Jagatramka - PLAINTIFF)

  text	                                               |             entities
0	Arun Kumar Jagatramka vs Ultrabulk AS on 22 Se...	  | [Arun Kumar Jagatramka - PLAINTIFF, Ultrabulk ...
1	Author Biren Vaishnav	                              |  [Biren Vaishnav - PERSON]
2	The Supreme Court ruled in favor of Jane Smith.	    |   [Supreme Court - LOC, Jane Smith - PLAINTIFF]
3	The Gujarat High Court issued a judgment in Ah...	  |  [Gujarat High Court - ORG, Ahmedabad - LOC]

API Documentation :

output for example only

2. install via Github

3. extract_DataFrame(df) >>

   ner = Custom_Ner_Dataset()
   data = ner.extract_DataFrame(df)
   

   output :

text	entities
0	Arun Kumar Jagatramka vs Ultrabulk AS on 22 Se...	[(0, 21, PLAINTIFF), (25, 37, Defender), (41, ...
1	Author Biren Vaishnav	[(7, 21, PERSON)]
2	The Supreme Court ruled in favor of Jane Smith.	[(4, 17, LOC), (36, 46, PLAINTIFF)]
3	The Gujarat High Court issued a judgment in Ah...	[(4, 22, ORG), (44, 53, LOC)]

2. to_dataset(data) >>

   import pandas as pd
   import numpy as np
   df = pd.DataFrame(ner.to_dataset(data))
   

   output :

         id	                                                     tokens	ner_tags
    0	0	[Arun, Kumar, Jagatramka, vs, Ultrabulk, AS, o...	 [B-PLAINTIFF, I-PLAINTIFF, I-PLAINTIFF, O, B-D...
    1	1	[Author, Biren, Vaishnav]	[O, B-PERSON, I-PERSON]
    2	8	[The, Supreme, Court, ruled, in, favor, of, Ja...	  [O, B-LOC, I-LOC, O, O, O, O, B-PLAINTIFF, I-P...
    3	9	[The, Gujarat, High, Court, issued, a, judgmen...	  [O, B-ORG, I-ORG, I-ORG, O, O, O, O, B-LOC, O]
   

3. Create _label_maps to create Huggingface Dataset :

```python
labels = []
for i in df["ner_tags"].tolist():
  labels.extend(i)
labels = np.unique(labels).tolist()
labels
```

output :

   ['B-DATE',
 'B-Defender',
 'B-LOC',
 'B-ORG',
 'B-PERSON',
 'B-PLAINTIFF',
 'I-DATE',
 'I-Defender',
 'I-LOC',
 'I-ORG',
 'I-PERSON',
 'I-PLAINTIFF',
 'O']

5. to_huggingface_dataset(data,labels) >>

   dataset = ner.to_huggingface_dataset(df,labels)
   dataset = dataset.train_test_split(test_size=0.1)
   dataset
   

   output :

    DatasetDict({
   train: Dataset({
       features: ['id', 'tokens', 'ner_tags'],
       num_rows: 3
   })
   test: Dataset({
       features: ['id', 'tokens', 'ner_tags'],
       num_rows: 1
   })
   

   })

6. coreference_model(text) >>>

   ner.coreference_model(text:str)  
   

   input : text = "John is Victim. He is Innocent"
   output : He mentions John it returns in json format which text,mentions,and span ...