ium_426206/dlgssdpytorch copy.py
2021-05-07 20:16:31 +02:00

180 lines
5.8 KiB
Python

import zipfile
import torch
import pandas as pd
import matplotlib
import matplotlib.pyplot as plt
import datetime
import numpy as np
from kaggle.api.kaggle_api_extended import KaggleApi
import torch.nn as nn
import torch.optim as optim
from torch.utils.data.dataset import random_split
from torch.utils.data import Dataset, TensorDataset, DataLoader
from torchviz import make_dot
from sklearn import preprocessing
# api = KaggleApi()
# api.authenticate()
# api.dataset_download_file('apoorvaappz/global-super-store-dataset',
# file_name='Global_Superstore2.csv', path='./')
# with zipfile.ZipFile('Global_Superstore2.csv.zip', 'r') as zipref:
# zipref.extractall('.')
data = pd.read_csv("Global_Superstore2.csv", header=0, sep=',')
data["Order Date"] = pd.to_datetime(data["Order Date"])
data = data.sort_values(by="Order Date")
#print(data)
byMonthsYears = {}
for index, row in data.iterrows():
#datee = datetime.datetime.strptime(row['Order Date'], "%d-%m-%Y")
#byMonthsYears.setdefault(datee.strftime("%m-%Y"), 0)
#byMonthsYears[datee.strftime("%m-%Y")] += row['Sales']
byMonthsYears.setdefault(row['Order Date'].strftime("%d-%m-%Y"), 0)
byMonthsYears[row['Order Date'].strftime("%d-%m-%Y")] += row['Sales']
df = data.groupby('Order Date').agg({'Customer Name':'count', 'Sales': 'sum'}).reset_index().rename(columns={'Sales':'Sales sum', 'Customer Name':'Sales count'})
#normalizacja danych
flcols = df[['Sales count', 'Sales sum']].columns
x = df[['Sales count', 'Sales sum']].values
# min_max_scaler = preprocessing.MinMaxScaler()
max_abs_scaler = preprocessing.MaxAbsScaler()
# x_scaled = min_max_scaler.fit_transform(x)
x_scaled = max_abs_scaler.fit_transform(x)
normcols = pd.DataFrame(x_scaled, columns=flcols)
for col in flcols:
df[col] = normcols[col]
df.to_csv('mms_norm.csv')
exit()
# fig, ax = plt.subplots()
# fig.set_figheight(15)
# fig.set_figwidth(20)
# ax.scatter(df['Month and Year'], df['Sum of sales'])
#plt.show()
# # Data Generation
# np.random.seed(42)
# x = np.random.rand(100, 1)
# y = 1 + 2 * x + .1 * np.random.randn(100, 1)
# # Shuffles the indices
# idx = np.arange(100)
# np.random.shuffle(idx)
# # Uses first 80 random indices for train
# train_idx = idx[:80]
# # Uses the remaining indices for validation
# val_idx = idx[80:]
# # Generates train and validation sets
# x_train, y_train = x[train_idx], y[train_idx]
# x_val, y_val = x[val_idx], y[val_idx]
# x_tensor = torch.from_numpy(x_train).float()
# y_tensor = torch.from_numpy(y_train).float()
x_tensor = torch.tensor(df['Sales sum'].values).float()
y_tensor = torch.tensor(df['Sales count'].values).float()
dataset = TensorDataset(x_tensor, y_tensor)
#torch.manual_seed(42)
lengths = [int(len(dataset)*0.8), int(len(dataset)*0.2)]
train_dataset, val_dataset = random_split(dataset, lengths)
train_loader = DataLoader(dataset=train_dataset)
val_loader = DataLoader(dataset=val_dataset)
class LayerLinearRegression(nn.Module):
def __init__(self):
super().__init__()
# Instead of our custom parameters, we use a Linear layer with single input and single output
self.linear = nn.Linear(1, 1)
def forward(self, x):
# Now it only takes a call to the layer to make predictions
return self.linear(x)
model = LayerLinearRegression()
# Checks model's parameters
#print(model.state_dict())
lr = 1e-3
n_epochs = 100
loss_fn = nn.MSELoss(reduction='mean')
optimizer = optim.SGD(model.parameters(), lr=lr)
def make_train_step(model, loss_fn, optimizer):
# Builds function that performs a step in the train loop
def train_step(x, y):
# Sets model to TRAIN mode
model.train()
# Makes predictions
yhat = model(x)
# Computes loss
loss = loss_fn(y, yhat)
# Computes gradients
loss.backward()
# Updates parameters and zeroes gradients
optimizer.step()
optimizer.zero_grad()
# Returns the loss
return loss.item()
# Returns the function that will be called inside the train loop
return train_step
# Creates the train_step function for our model, loss function and optimizer
train_step = make_train_step(model, loss_fn, optimizer)
training_losses = []
validation_losses = []
print(model.state_dict())
# For each epoch...
for epoch in range(n_epochs):
losses = []
# Uses loader to fetch one mini-batch for training
for x_batch, y_batch in train_loader:
# NOW, sends the mini-batch data to the device
# so it matches location of the MODEL
# x_batch = x_batch.to(device)
# y_batch = y_batch.to(device)
# One stpe of training
loss = train_step(x_batch, y_batch)
losses.append(loss)
training_loss = np.mean(losses)
training_losses.append(training_loss)
# After finishing training steps for all mini-batches,
# it is time for evaluation!
# We tell PyTorch to NOT use autograd...
# Do you remember why?
with torch.no_grad():
val_losses = []
# Uses loader to fetch one mini-batch for validation
for x_val, y_val in val_loader:
# Again, sends data to same device as model
# x_val = x_val.to(device)
# y_val = y_val.to(device)
# What is that?!
model.eval()
# Makes predictions
yhat = model(x_val)
# Computes validation loss
val_loss = loss_fn(y_val, yhat)
val_losses.append(val_loss.item())
validation_loss = np.mean(val_losses)
validation_losses.append(validation_loss)
print(f"[{epoch+1}] Training loss: {training_loss:.3f}\t Validation loss: {validation_loss:.3f}")
# Checks model's parameters
print(model.state_dict())
print(np.mean(losses))
print(np.mean(val_losses))