130 lines
4.0 KiB
Python
130 lines
4.0 KiB
Python
#!/usr/bin/python
|
|
|
|
import numpy as np
|
|
import torch
|
|
from torch import nn
|
|
from torch.autograd import Variable
|
|
from sklearn.datasets import load_iris
|
|
from sklearn.model_selection import train_test_split
|
|
from sklearn.metrics import accuracy_score
|
|
import torch.nn.functional as F
|
|
import pandas as pd
|
|
from sklearn import preprocessing
|
|
import sys
|
|
|
|
|
|
class Model(nn.Module):
|
|
def __init__(self, input_dim):
|
|
super(Model, self).__init__()
|
|
self.layer1 = nn.Linear(input_dim, 100)
|
|
self.layer2 = nn.Linear(100, 60)
|
|
self.layer3 = nn.Linear(60, 5)
|
|
|
|
def forward(self, x):
|
|
x = F.relu(self.layer1(x))
|
|
x = F.relu(self.layer2(x))
|
|
x = F.softmax(self.layer3(x)) # To check with the loss function
|
|
return x
|
|
|
|
|
|
def load_dataset_raw():
|
|
""" Load data from .csv file. """
|
|
cars = pd.read_csv('./Car_Prices_Poland_Kaggle.csv', usecols=[1, 4, 5, 6, 10], sep=',')
|
|
return cars
|
|
|
|
|
|
def remove_rows(dataset):
|
|
# dataset.drop(dataset[dataset['mark'] == 'alfa-romeo'].index, inplace=True)
|
|
# dataset.drop(dataset[dataset['mark'] == 'chevrolet'].index, inplace=True)
|
|
# dataset.drop(dataset[dataset['mark'] == 'mitsubishi'].index, inplace=True)
|
|
# dataset.drop(dataset[dataset['mark'] == 'mini'].index, inplace=True)
|
|
# audi bmw ford opel volkswagen
|
|
|
|
new_data = dataset.loc[(dataset['mark'] == 'audi') | (dataset['mark'] == 'bmw') | (dataset['mark'] == 'ford') | (dataset['mark'] == 'opel') | (dataset['mark'] == 'volkswagen')]
|
|
return new_data
|
|
# dataset = dataset.drop(dataset)
|
|
# return dataset
|
|
|
|
|
|
def prepare_dataset_raw(dataset):
|
|
""" Label make column"""
|
|
le = preprocessing.LabelEncoder()
|
|
mark_column = np.array(dataset[:]['mark'])
|
|
le.fit(mark_column)
|
|
|
|
print(list(le.classes_))
|
|
lab = le.transform(mark_column)
|
|
feat = dataset.drop(['mark'], axis=1).to_numpy()
|
|
|
|
mm_scaler = preprocessing.MinMaxScaler()
|
|
feat = mm_scaler.fit_transform(feat)
|
|
|
|
return lab, feat
|
|
|
|
|
|
# def draw_plot(lbl):
|
|
# need to import matplotlib to work
|
|
# plt.hist(lbl, bins=[i for i in range(len(set(lbl)))], edgecolor="black")
|
|
# plt.xticks(np.arange(0, len(set(lbl)), 1))
|
|
# plt.show()
|
|
|
|
# Prepare dataset
|
|
print("Loading dataset...")
|
|
dataset = load_dataset_raw()
|
|
print("Dataset loaded")
|
|
|
|
print("Preparing dataset...")
|
|
dataset = remove_rows(dataset)
|
|
labels, features = prepare_dataset_raw(dataset)
|
|
print("Dataset prepared")
|
|
|
|
|
|
features_train, features_test, labels_train, labels_test = train_test_split(features, labels, random_state=42,
|
|
shuffle=True)
|
|
# Training
|
|
model = Model(features_train.shape[1])
|
|
optimizer = torch.optim.Adam(model.parameters(), lr=0.01)
|
|
loss_fn = nn.CrossEntropyLoss()
|
|
|
|
# number of epochs is parametrized
|
|
try:
|
|
epochs = int(sys.argv[1])
|
|
except Exception as e:
|
|
print(e)
|
|
print("Setting default epochs value to 1000.")
|
|
epochs = 1000
|
|
|
|
print(f"Number of epochs: {epochs}")
|
|
|
|
print("Starting model training...")
|
|
x_train, y_train = Variable(torch.from_numpy(features_train)).float(), Variable(torch.from_numpy(labels_train)).long()
|
|
for epoch in range(1, epochs + 1):
|
|
print("Epoch #", epoch)
|
|
y_pred = model(x_train)
|
|
loss = loss_fn(y_pred, y_train)
|
|
print(f"The loss calculated: {loss}")
|
|
|
|
# Zero gradients
|
|
optimizer.zero_grad()
|
|
loss.backward() # Gradients
|
|
optimizer.step() # Update
|
|
print("Model training finished")
|
|
|
|
x_test = Variable(torch.from_numpy(features_test)).float()
|
|
pred = model(x_test)
|
|
pred = pred.detach().numpy()
|
|
print("The accuracy is", accuracy_score(labels_test, np.argmax(pred, axis=1)))
|
|
|
|
# Checking for first value
|
|
# print(np.argmax(model(x_test[0]).detach().numpy(), axis=0))
|
|
# print(labels_test[0])
|
|
|
|
print("Saving model to file...")
|
|
torch.save(model, "CarPrices_pytorch_model.pkl")
|
|
print("Model saved with name: CarPrices_pytorch_model.pkl")
|
|
|
|
saved_model = torch.load("CarPrices_pytorch_model.pkl")
|
|
print(np.argmax(saved_model(x_test[0]).detach().numpy(), axis=0))
|
|
|
|
pd_predictions = pd.DataFrame(pred)
|
|
pd_predictions.to_csv("./prediction_results.csv") |