license: apache-2.0
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- cnn
- amazon_reviews
datasets:
- yassiracharki/Amazon_Reviews_Binary_for_Sentiment_Analysis
language:
- en
library_name: fasttext
Model Card for Model ID
Downloads
!pip install contractions !pip install textsearch !pip install tqdm
import nltk nltk.download('punkt')
Fundamental classes
import tensorflow as tf from tensorflow import keras import pandas as pd import numpy as np
Time
import time import datetime
Preprocessing
from tensorflow.keras.preprocessing.text import Tokenizer from tensorflow.keras.preprocessing import sequence from sklearn.preprocessing import LabelEncoder import contractions from bs4 import BeautifulSoup import re import tqdm import unicodedata
seed = 3541 np.random.seed(seed)
Define a dummy loss to bypass the error during model loading
def dummy_loss(y_true, y_pred): return tf.reduce_mean(y_pred - y_true)
Loading the model Trained on Amazon reviews
modelAmazon = keras.models.load_model( '/kaggle/input/pre-trained-model-binary-cnn-nlp-amazon-reviews/tensorflow1/pre_trained_sentiment_analysis_cnn_model_amazon_reviews/1/Binary_Classification_86_Amazon_Reviews_CNN.h5', compile=False )
Compile the model with the correct loss function and reduction
modelAmazon.compile( optimizer='adam', loss=keras.losses.BinaryCrossentropy(reduction=tf.keras.losses.Reduction.SUM_OVER_BATCH_SIZE), metrics=['accuracy'] )
Loading Amazon test data
dataset_test_Amazon = pd.read_csv('/kaggle/input/amazon-reviews-for-sa-binary-negative-positive-csv/amazon_review_sa_binary_csv/test.csv')
Loading Amazon train data (to be used on the label encoder)
dataset_train_Amazon = pd.read_csv('/kaggle/input/amazon-reviews-for-sa-binary-negative-positive-csv/amazon_review_sa_binary_csv/train.csv')
Shuffling the Test Data
test_Amazon = dataset_test_Amazon.sample(frac=1) train_Amazon = dataset_train_Amazon.sample(frac=1)
Taking a tiny portion of the database (because it will only be used on the label encoder)
train_Amazon = dataset_train_Amazon.iloc[:100, :]
Taking only necessary columns
y_test_Amazon = test_Amazon['class_index'].values X_train_Amazon = train_Amazon['review_text'].values y_train_Amazon = train_Amazon['class_index'].values
Preprocess corpus function
def pre_process_corpus(corpus): processed_corpus = [] for doc in tqdm.tqdm(corpus): doc = contractions.fix(doc) doc = BeautifulSoup(doc, "html.parser").get_text() doc = unicodedata.normalize('NFKD', doc).encode('ascii', 'ignore').decode('utf-8', 'ignore') doc = re.sub(r'[^a-zA-Z\s]', '', doc, re.I|re.A) doc = doc.lower() doc = doc.strip() processed_corpus.append(doc) return processed_corpus
Preprocessing the Data
X_test_Amazon = pre_process_corpus(test_Amazon['review_text'].values) X_train_Amazon = pre_process_corpus(X_train_Amazon)
Creating and Fitting the Tokenizer
etc ...
More info on the Model's page on Kaggle :
https://www.kaggle.com/models/yacharki/pre-trained-model-binary-cnn-nlp-amazon-reviews