predict.py

from ast import parse
import os
from argparse import ArgumentParser
import tensorflow as tf
import pandas as pd
from tensorflow.python.keras.layers import embeddings 
from data import *
from constant import *
import numpy as np
from tensorflow.keras.preprocessing.sequence import pad_sequences
from constant import *
if __name__ == "__main__":
    parser = ArgumentParser()
    home_dir = os.getcwd() 
    parser.add_argument("--smallCharCNN-folder", default="smallCharCNN", type=str)
    parser.add_argument("--vocab-folder", default= '{}/saved_vocab/CharCNN/'.format(home_dir), type= str)
    parser.add_argument("--largeCharCNN-folder", default="largeCharCNN", type=str)
    parser.add_argument("--test-file", default="test.csv", type=str)
    parser.add_argument("--model", default="small", type=str)
    parser.add_argument("--result-file", default="result.csv", type=str)
    args = parser.parse_args()

    print('---------------------Welcome to CharCNN-------------------')
    print("Team Leader")
    print("1. Github: hoangcaobao")
    print("Team member")
    print('1. Github: Nguyendat-bit')
    print('2. Github: aestheteeism')
    print('---------------------------------------------------------------------')
    print('Training CharCNN model with hyper-params:') 
    for i, arg in enumerate(vars(args)):
        print('{}. {}: {}'.format(i, arg, vars(args)[arg]))
    print('===========================')

    # Loading Models 
    # Small-CharCNN model
    if(args.model=="small"):
        model = tf.keras.models.load_model(args.smallCharCNN_folder)
    # Large-CharCNN model
    else:
        model = tf.keras.models.load_model(args.largeCharCNN_folder)

    # Loading Tokenizer
    print('=============Loading Tokenizer================')
    print('Begin...')
    dataset = Dataset(vocab_folder= args.vocab_folder)

    label_dict = dataset.label_dict
    print('Done!!!')

    # Load test sentences 
    sentence = pd.read_csv(args.test_file)
    sentence.columns = ['sentence']
    sentence = np.array(sentence['sentence'])
    test = [dataset.preprocess_data(i) for i in sentence]
    test = dataset.tokenizer.texts_to_sequences(test)
    test = pad_sequences(test, maxlen= model.layers[0].input_length, padding=padding)
    
    # Preidct
    print("================Predicting================")
    predict=model.predict(test)
    predict=np.argmax(predict, axis=1)
    predict=predict.astype("O")

    #Decode predict
    label_dict = dict((i,l) for l,i in label_dict.items())
    for label_num in label_dict.keys():
        predict[predict == label_num] = label_dict[label_num]
    #Save to csv
    data=np.column_stack((sentence, predict))
    df=pd.DataFrame(data, columns=['sentence', 'label'])
    df.to_csv(args.result_file, index=False)
    print("End Predicting. Now your result will be in {}".format(args.result_file))