def preprocessing(chunk_size=100):
# load data
loaders, encoder = createLoaders(extras=extras, chunk_size=chunk_size)
dataloaders = dict(zip(['train', 'val', 'test'], loaders))
print('------- Info ---------')
for phase in dataloaders:
print('- %s size: %i' % (phase, len(dataloaders[phase])))
print('----------------------')
return dataloaders, encoder
import utils as ut
import torch
from torch.utils.data import DataLoader
import torch.nn as nn
from torch.nn import functional as F
from torch.autograd import Variable
from torch import optim
from music_dataloader import createLoaders
import numpy as np
# Check if your system supports CUDA
use_cuda = torch.cuda.is_available()
# Setup GPU optimization if CUDA is supported
if use_cuda:
computing_device = torch.device("cuda")
extras = {"num_workers": 1, "pin_memory": True}
print("CUDA is supported")
else: # Otherwise, train on the CPU
computing_device = torch.device("cpu")
extras = False
print("CUDA NOT supported")
# load data
train_loader, val_loader, test_loader = createLoaders(extras=extras)
criterion = nn.CrossEntropyLoss()
# Check if your system supports CUDA
use_cuda = torch.cuda.is_available()
# Setup GPU optimization if CUDA is supported
if use_cuda:
computing_device = torch.device("cuda")
extras = {"num_workers": 1, "pin_memory": True}
print("CUDA is supported")
else: # Otherwise, train on the CPU
computing_device = torch.device("cpu")
extras = False
print("CUDA NOT supported")
# load data
train_loader, val_loader, test_loader, one_hot_length = createLoaders(
extras=extras)
RNN = rnn.RNN(hidden_size, one_hot_length, computing_device)
RNN = RNN.to(computing_device)
print("Model on CUDA?", next(RNN.parameters()).is_cuda)
print("Model on CUDA?",
next(RNN.parameters()).is_cuda,
file=open("output.txt", "a"))
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(RNN.parameters())
# Track the loss across training
chunk_train_loss = []
# calculate training and validation loss per N times through the whole training process
batch_train_loss = []