| # Download nltk and numpy | |
| import os | |
| os.system('pip install nltk numpy') | |
| import nltk | |
| import numpy as np | |
| nltk.download('punkt') | |
| def train_model(corpus): | |
| tokens = nltk.word_tokenize(corpus) | |
| model = {} | |
| for i in range(len(tokens) - 1): | |
| if tokens[i] in model: | |
| model[tokens[i]].append(tokens[i + 1]) | |
| else: | |
| model[tokens[i]] = [tokens[i + 1]] | |
| return model | |
| import pickle | |
| # Train the model on a given corpus | |
| corpus = open('corpus.txt').read() | |
| model = train_model(corpus) | |
| # Save the model to a file | |
| with open("model.pkl", "wb") as f: | |
| pickle.dump(model, f) | |