"""
Callback Extension Example
================================
This example should illustrate how to extend the training
using simple callbacks. In particular we will modulate the learning rate
with a sawtooth function and clip the gradients by value
"""
##################################################
# change directories to your needs
from inferno.trainers.callbacks.logging.tensorboard import TensorboardLogger
from inferno.utils.python_utils import ensure_dir
LOG_DIRECTORY = ensure_dir('log/sawtooth')
SAVE_DIRECTORY = ensure_dir('save')
DATASET_DIRECTORY = ensure_dir('dataset')
print("\n \n LOGDIR", LOG_DIRECTORY)
##################################################
# shall models be downloaded
DOWNLOAD_CIFAR = True
USE_CUDA = True
##################################################
# Build torch model
import torch.nn as nn
from inferno.extensions.layers import ConvELU2D
from inferno.extensions.layers import Flatten
model = nn.Sequential(
ConvELU2D(in_channels=3, out_channels=256, kernel_size=3),
nn.MaxPool2d(kernel_size=2, stride=2),
from inferno.trainers.callbacks.logging.tensorboard import TensorboardLogger
import torch
import torch.nn as nn
import torch.optim as optim
import torch.utils.data
import numpy as np
import torch.nn.functional as F
import torchvision.transforms as transforms
import warnings
warnings.filterwarnings("ignore", category=UserWarning)
from inferno.utils.python_utils import ensure_dir
LOG_DIRECTORY = ensure_dir('./logs_2')
BATCHSIZE = 8
N_DIRECTIONS = 8
# unsq = transforms.Lambda(lambda x: torch.unsqueeze(x, 0))
transpose = transforms.Lambda(lambda x: torch.transpose(x, 0, 1))
squeeze = transforms.Lambda(lambda x: torch.squeeze(x, 1))
fromnumpy = transforms.Lambda(lambda x: torch.from_numpy(x))
trans = transforms.Compose([fromnumpy])
trans2 = transforms.Compose([fromnumpy, squeeze])
imageset_train = HDF5VolumeLoader(path='./train-volume.h5', path_in_h5_dataset='data',
transforms=trans, **yaml2dict('config_train.yml')['slicing_config'])