38 lines
965 B
Python
38 lines
965 B
Python
from transformers import AutoTokenizer
|
|
from transformers import pipeline
|
|
import re
|
|
|
|
model = 'application/models/sentiment_model'
|
|
tokenizer = AutoTokenizer.from_pretrained('application/tokenizers/sentiment_tokenizer')
|
|
|
|
def sentiment_prediction(data):
|
|
pipe = pipeline('text-classification', model=model, tokenizer = tokenizer)
|
|
result = pipe(data)
|
|
|
|
return result
|
|
|
|
def clear_data(data):
|
|
data = [re.sub(r"[^A-Za-zżźćńółęąśŻŹĆĄŚĘŁÓŃ ']+", r"", i) for i in data['sentences']]
|
|
data = [i.strip() for i in data]
|
|
data = [i.lower() for i in data]
|
|
return data
|
|
|
|
def count_predictions(predictions):
|
|
l0 = 0
|
|
l1 = 0
|
|
l2 = 0
|
|
all = {}
|
|
|
|
for i in predictions:
|
|
if i['label'] == 'LABEL_0':
|
|
l0 += 1
|
|
if i['label'] == 'LABEL_1':
|
|
l1 += 1
|
|
if i['label'] == 'LABEL_2':
|
|
l2 += 1
|
|
|
|
all['positive'] = l1
|
|
all['negative'] = l0
|
|
all['neutral'] = l2
|
|
|
|
return all |