wtberry

import pandas as pd 
import os 
import numpy as np
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.model_selection import train_test_split

# setting up path to the data file
PATH = os.path.dirname(os.path.realpath(__file__))
PATH = os.path.join(PATH, 'data')
print(PATH)

wtberry

wtberry

from sklearn.preprocessing import OrdinaryEncoder

# creating mapping from unique label texts to unique integers
# note this can be re-used to encode and decode the labels after as well
encoder = OrdinaryEncoder().fit(df['code'])

# using the encoder to encode the entire dataset
y = encoder.transform(encoder)

wtberry

from sklearn.feature_extraction.text import CountVectorizer

# Initialize and fit CountVectorizer with given text documents
vectorizer = CountVectorizer().fit(df['name'])

# use the vectorizer to transform the document into word count vectors (Sparse)
word_mat = vectorizer.transform(df['name'])

wtberry

from sklearn.naive_bayes import MultinomialNB

# instantiate the model as clf(classifier) and train it
clf = MultinomialNB()
clf.fit(x_train, y_train)

wtberry

from sklearn.model_selection import train_test_split

x_train, x_test, y_train, y_test = train_test_split(word_mat, y, test_size=0.3)

wtberry

from sklearn.model_selection import train_test_split

x_train, x_test, y_train, y_test = train_test_split(word_mat, y, test_size=0.3)

wtberry

wtberry

wtberry