Files
drift/data_loader/pytorch_dataset.py
T
Daniel Szemerey ee35332f58 feature(Models): Implemented a basic Neural Network with Pytorch-Lightning (#101)
* feat: Added base functions for Neural Net.

* feat: Added function to handle Neural Nets.

* fix: Fixed fit loop

* feat: Neural Net trains now, need to test it.

* feat: Prediction now works on the neural net.

* fix: Put back config and run_pipeline.py

* fix: Took out import from run_pipeline.

* fix(Models): added get_name(), adjusted pytorch model output size

* fix(Tests): fixed tests

Co-authored-by: Mark Aron Szulyovszky <mark.szulyovszky@gmail.com>
2022-01-05 12:25:03 +01:00

26 lines
681 B
Python

import os
import pandas as pd
import torch
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
import numpy as np
class TimeSeriesDataset(Dataset):
def __init__(self, X: np.ndarray, y: np.ndarray):
self.X = X
self.y = y
def __len__(self):
return len(self.X)
def __getitem__(self, idx):
return self.X[idx].astype(float), self.y[idx].astype(float)
def get_dataloader(X: np.ndarray, y: np.ndarray, batch_size: int = 32, shuffle: bool = True):
training_data = TimeSeriesDataset(X, y)
train_dataloader = DataLoader(training_data, batch_size=batch_size, shuffle=shuffle)
return train_dataloader