sentiment_CNN.py

# -*- coding: utf-8 -*-
"""Untitled15.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1yAnc6o7AXzZeWtD4xUraYMgwzchFtwmU
"""

import numpy as np
import pandas as pd
import seaborn as sns

df=pd.read_csv('/content/updated_sentiment_analysis(2).csv')

df.head()

df=df.sample(10000)
data=df

from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences

tokenizer = Tokenizer(num_words=500, split=' ')
tokenizer.fit_on_texts(df['review'].values)
x = tokenizer.texts_to_sequences(df['review'])

x = pad_sequences(x, maxlen=80)
y = df['sentiment_positive']

max_words=500



from sklearn.model_selection import train_test_split

x_train,x_test,y_train,y_test = train_test_split(x,y,random_state=2021,test_size=0.1)

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense,Flatten

from keras.layers.embeddings import Embedding

from keras.layers.convolutional import Conv1D
from keras.layers.convolutional import MaxPooling1D

cnn_model=Sequential()

cnn_model.add(Embedding(1000, 300, input_length=max_words))
cnn_model.add(Conv1D(64,3,activation='relu'))
cnn_model.add(MaxPooling1D(pool_size=2))
cnn_model.add(Conv1D(32,3,activation='relu'))
cnn_model.add(MaxPooling1D(pool_size=2))
cnn_model.add(Dense(8, activation='relu'))
cnn_model.add(Dense(1, activation='sigmoid'))

cnn_model.compile(optimizer='adam', loss="binary_crossentropy", metrics=['accuracy'])

cnn_model.summary()

cnn_model.fit(x_train, y_train, validation_data=(x_test, y_test), epochs=50, batch_size=128, verbose=2)