Update app.py
Browse files
app.py
CHANGED
|
@@ -1,17 +1,33 @@
|
|
| 1 |
-
import gradio as gr
|
| 2 |
-
import joblib
|
| 3 |
import os
|
|
|
|
|
|
|
|
|
|
| 4 |
import spacy
|
| 5 |
import nltk
|
| 6 |
from nltk.stem import WordNetLemmatizer
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
-
|
| 12 |
-
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
label_mapping = {
|
| 16 |
0: 'sadness',
|
| 17 |
1: 'joy',
|
|
@@ -21,26 +37,39 @@ label_mapping = {
|
|
| 21 |
5: 'surprise'
|
| 22 |
}
|
| 23 |
|
| 24 |
-
#
|
| 25 |
-
|
| 26 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
|
|
|
|
| 29 |
def predict_emotion(text):
|
| 30 |
cleaned_text = fully_cleaned(text)
|
| 31 |
-
|
| 32 |
text_tfidf = tfidf_vectorizer.transform([cleaned_text])
|
| 33 |
prediction = logistic_regression_tfidf_model.predict(text_tfidf)[0]
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
return predicted_emotion
|
| 37 |
|
| 38 |
-
#
|
| 39 |
-
iface = gr.Interface(
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
|
|
|
|
|
|
| 44 |
|
| 45 |
# Launch the interface
|
| 46 |
-
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
import os
|
| 2 |
+
import re
|
| 3 |
+
import joblib
|
| 4 |
+
import gradio as gr
|
| 5 |
import spacy
|
| 6 |
import nltk
|
| 7 |
from nltk.stem import WordNetLemmatizer
|
| 8 |
+
from spacy.lang.en.stop_words import STOP_WORDS
|
| 9 |
+
|
| 10 |
+
# 1. Setup NLTK and SpaCy dependencies
|
| 11 |
+
try:
|
| 12 |
+
nltk.data.find('tokenizers/punkt')
|
| 13 |
+
nltk.data.find('tokenizers/punkt_tab')
|
| 14 |
+
except LookupError:
|
| 15 |
+
nltk.download('punkt')
|
| 16 |
+
nltk.download('punkt_tab')
|
| 17 |
|
| 18 |
+
try:
|
| 19 |
+
nlp = spacy.load("en_core_web_sm")
|
| 20 |
+
except OSError:
|
| 21 |
+
from spacy.cli import download
|
| 22 |
+
download("en_core_web_sm")
|
| 23 |
|
| 24 |
+
lemmatizer = WordNetLemmatizer()
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
logistic_regression_tfidf_model = joblib.load('logistic_regression_bow.pkl')
|
| 28 |
+
tfidf_vectorizer = joblib.load('vectorizer_tfidf.pkl')
|
| 29 |
+
|
| 30 |
+
# 3. Label mapping
|
| 31 |
label_mapping = {
|
| 32 |
0: 'sadness',
|
| 33 |
1: 'joy',
|
|
|
|
| 37 |
5: 'surprise'
|
| 38 |
}
|
| 39 |
|
| 40 |
+
# 4. Include the text cleaning functions from your notebook
|
| 41 |
+
def basic_cleaning(text):
|
| 42 |
+
text = re.sub(r'http\S+', '', text)
|
| 43 |
+
text = re.sub(r'[^a-zA-Z\s]', '', text)
|
| 44 |
+
text = re.sub(r'\w*\d\w*', '', text)
|
| 45 |
+
text = re.sub(r'<.*?>', '', text)
|
| 46 |
+
text = text.lower()
|
| 47 |
+
return text
|
| 48 |
|
| 49 |
+
def fully_cleaned(text):
|
| 50 |
+
cleaned = basic_cleaning(text)
|
| 51 |
+
doc = nlp(cleaned)
|
| 52 |
+
tokens = [token.text for token in doc]
|
| 53 |
+
tokens = [token for token in tokens if token not in STOP_WORDS]
|
| 54 |
+
tokens = [lemmatizer.lemmatize(token) for token in tokens]
|
| 55 |
+
return ' '.join(tokens)
|
| 56 |
|
| 57 |
+
# 5. Prediction function
|
| 58 |
def predict_emotion(text):
|
| 59 |
cleaned_text = fully_cleaned(text)
|
|
|
|
| 60 |
text_tfidf = tfidf_vectorizer.transform([cleaned_text])
|
| 61 |
prediction = logistic_regression_tfidf_model.predict(text_tfidf)[0]
|
| 62 |
+
return label_mapping.get(prediction, "Unknown")
|
|
|
|
|
|
|
| 63 |
|
| 64 |
+
# 6. Gradio Interface
|
| 65 |
+
iface = gr.Interface(
|
| 66 |
+
fn=predict_emotion,
|
| 67 |
+
inputs=gr.Textbox(lines=5, placeholder="Enter text here..."),
|
| 68 |
+
outputs="text",
|
| 69 |
+
title="Emotion Classifier",
|
| 70 |
+
description="Enter a sentence to classify its emotion (sadness, joy, love, anger, fear, surprise)."
|
| 71 |
+
)
|
| 72 |
|
| 73 |
# Launch the interface
|
| 74 |
+
if __name__ == "__main__":
|
| 75 |
+
iface.launch()
|