File size: 3,760 Bytes
4a6159c 7de7bf4 4a6159c 570b6e4 4a6159c 570b6e4 4a6159c 7de7bf4 4a6159c 7de7bf4 4a6159c 7de7bf4 1a4b146 4a6159c 1a4b146 3f0df44 570b6e4 4a6159c 2caced7 4a6159c 2caced7 4a6159c 1d3978a 4a6159c 1d3978a 4a6159c 7de7bf4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 |
from haystack.nodes import TransformersDocumentClassifier
from haystack.schema import Document
from typing import List, Tuple
import configparser
import streamlit as st
from pandas import DataFrame, Series
import logging
from utils.preprocessing import processingpipeline
config = configparser.ConfigParser()
config.read_file(open('paramconfig.cfg'))
@st.cache(allow_output_mutation=True)
def load_sdgClassifier():
"""
loads the document classifier using haystack, where the name/path of model
in HF-hub as string is used to fetch the model object.
1. https://docs.haystack.deepset.ai/reference/document-classifier-api
2. https://docs.haystack.deepset.ai/docs/document_classifier
Return: document classifier model
"""
logging.info("Loading classifier")
doc_classifier_model = config.get('sdg','MODEL')
doc_classifier = TransformersDocumentClassifier(
model_name_or_path=doc_classifier_model,
task="text-classification")
return doc_classifier
def sdg_classification(haystackdoc:List[Document])->Tuple[DataFrame,Series]:
"""
Text-Classification on the list of texts provided. Classifier provides the
most appropriate label for each text. these labels are in terms of if text
belongs to which particular Sustainable Devleopment Goal (SDG).
Params
---------
haystackdoc: List of haystack Documents. The output of Preprocessing Pipeline
contains the list of paragraphs in different format,here the list of
Haystack Documents is used.
Returns
----------
df: Dataframe with two columns['SDG:int', 'text']
x: Series object with the unique SDG covered in the document uploaded and
the number of times it is covered/discussed/count_of_paragraphs.
"""
logging.info("running SDG classifiication")
threshold = float(config.get('sdg','THRESHOLD'))
classifier = load_sdgClassifier()
results = classifier.predict(haystackdoc)
labels_= [(l.meta['classification']['label'],
l.meta['classification']['score'],l.content,) for l in results]
df = DataFrame(labels_, columns=["SDG","Relevancy","text"])
df = df.sort_values(by="Relevancy", ascending=False).reset_index(drop=True)
df.index += 1
df =df[df['Relevancy']>threshold]
x = df['SDG'].value_counts()
df= df.drop(['Relevancy'], axis = 1)
return df, x
def runSDGPreprocessingPipeline()->List[Document]:
"""
creates the pipeline and runs the preprocessing pipeline,
the params for pipeline are fetched from paramconfig
Return
--------------
List[Document]: When preprocessing pipeline is run, the output dictionary
has four objects. For the Haysatck implementation of SDG classification we,
need to use the List of Haystack Document, which can be fetched by
key = 'documents' on output.
"""
file_path = st.session_state['filepath']
file_name = st.session_state['filename']
sdg_processing_pipeline = processingpipeline()
split_by = config.get('sdg','SPLIT_BY')
split_length = int(config.get('sdg','SPLIT_LENGTH'))
split_overlap = int(config.get('sdg','SPLIT_OVERLAP'))
output_sdg_pre = sdg_processing_pipeline.run(file_paths = file_path,
params= {"FileConverter": {"file_path": file_path, \
"file_name": file_name},
"UdfPreProcessor": {"removePunc": False, \
"split_by": split_by, \
"split_length":split_length,\
"split_overlap": split_overlap}})
return output_sdg_pre['documents']
|