mbednarski

mbednarski

mbednarski

from pathlib import Path

class Dataset:
    def __init__(self):
        self.train_dir = Path('data/raw/aclImdb/train')
        self.test_dir = Path('data/raw/aclImdb/test')

    def _get_set(self, limit=None, directory):
        x = []
        y = []

mbednarski

import pickle

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import classification_report

from dataset import Dataset


dset = Dataset()

mbednarski

import pickle

with open('tfidf.pickle', 'rb') as f:
    tfidf = pickle.load(f)

with open('model.pickle', 'rb') as f:
    clf = pickle.load(f)

x = input("Please enter your phrase: ")
y = clf.predict_proba(tfidf.transform([x]))

mbednarski

# (...)

class Dataset:
    def __init__(self, train_dir='data/raw/aclImdb/train', test_dir='data/raw/aclImdb/test'):
        self.train_dir = Path(train_dir)
        self.test_dir = Path(test_dir)
       
# (...)

mbednarski

import pickle

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import classification_report


class DumbModel:
    def __init__(self, vocab_size=10_000):
        self.vocab_size = vocab_size

mbednarski

from setuptools import setup

setup(name='mlp',
      packages=['mlp'],
      version='0.0.1dev1',
      )

mbednarski

import os
from sklearn.metrics import classification_report

from mlp import DumbModel, Dataset

def train_model(dataset_dir, model_file, vocab_size):
    print(f'Training model from directory {dataset_dir}')
    print(f'Vocabulary size: {vocab_size}')

    train_dir = os.path.join(dataset_dir, 'train')

mbednarski

"""MLP - machine-learning-production

Usage:
    mlp.py train <dataset-dir> <model-file> [--vocab-size=<vocab-size>]
    mlp.py ask <model-file> <question>
    mlp.py (-h | --help)

Arguments:
    <dataset-dir>  Directory with dataset.
    <model-file>   Serialized model file.