Next Word Predictor
A Deep Learning Model to predict next word in a sequence using LSTM.
The Application is created using Tensorflow and Python.
Getting Started
- Clone the repo and cd into the directory
$ git clone https://github.com/raj713335/Next_Word_Predictor.git
$ cd Next_Word_PredictorThe training data is present in the corpus.txt file.
The WordCloud of the traing Data
Install tensorflow and all the other required libraries
$ pip install tensorflow, keras
import numpy as np
from nltk.tokenize import RegexpTokenizer
from keras.models import Sequential, load_model
from keras.layers import LSTM
from keras.layers.core import Dense, Activation
from keras.optimizers import RMSprop
import matplotlib.pyplot as plt
import pickle
import heapq
import tensorflow as tf
from wordcloud import WordCloud, STOPWORDS
import numpy as npy
from PIL import ImageNow To train the Model Enter to the NLP_DEEP_LEARNING FOLDER and then again to the sub directory CODE and run the Model Creator.py file
$ cd NLP_DEEP_LEARNING
$ cd CODE
$ python Model Creator.pyThe Summary of the model created
Model: "sequential_1"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
lstm_1 (LSTM) (None, 128) 749056
_________________________________________________________________
dense_1 (Dense) (None, 1334) 172086
_________________________________________________________________
activation_1 (Activation) (None, 1334) 0
=================================================================
Total params: 921,142
Trainable params: 921,142
Non-trainable params: 0
_________________________________________________________________
Now To run the Model keras_next_word_model.h5 run the Output.py file
$ python Output.pyAfter that in command line you can enter the input for which you want to check the output.
NOW TO RUN THE GUI VERSION OF THE WORD_PREDICTOR
$ cd ..
$ cd ..
$ cd GUI
$ python GUI_WP.py
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent