Kulikovpavel

ddd

Kulikovpavel

LOG_DIR = '"./drive/My Drive/log/"'
get_ipython().system_raw(
    'tensorboard --logdir {} --host 0.0.0.0 --port 6006 &'
    .format(LOG_DIR)
)
writer = SummaryWriter(log_dir='./drive/My Drive/log/runs/test-1')

Kulikovpavel

epochs = 5
steps = 0
running_loss = 0
print_every = 10
writer = SummaryWriter(log_dir='./log/runs/test-1')
for epoch in range(epochs):
    for inputs, labels in trainloader:
        steps += 1
  
        # Move input and label tensors to the default device

Kulikovpavel

# To GPU
model.to(device)
criterion = nn.CrossEntropyLoss()
# Only train the classifier parameters, feature parameters are frozen
optimizer = optim.Adam(model.classifier.parameters())

Kulikovpavel

num_classes = 2

model.classifier = nn.Sequential(
    nn.Dropout(p=0.1),
    nn.Conv2d(512, num_classes, kernel_size=(1, 1), stride=(1, 1)), 
    nn.ReLU(), 
    nn.AvgPool2d(kernel_size=13, stride=1, padding=0)
)

model.num_classes = num_classes

Kulikovpavel

# Use GPU if it's available
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = models.squeezenet1_1(pretrained=True)
# Freeze parameters so we don't backprop through them
for param in model.parameters():
    param.requires_grad = False

Kulikovpavel

data_dir = 'Cat_Dog_data'
# TODO: Define transforms for the training data and testing data
train_transforms = transforms.Compose([transforms.RandomRotation(30),
                                       transforms.RandomResizedCrop(224),
                                       transforms.RandomHorizontalFlip(),
                                       transforms.ToTensor(),
                                       transforms.Normalize([0.485, 0.456, 0.406],
                                                            [0.229, 0.224, 0.225])])
test_transforms = transforms.Compose([transforms.Resize(255),
                                      transforms.CenterCrop(224),

Kulikovpavel

!wget https://bin.equinox.io/c/4VmDzA7iaHb/ngrok-stable-linux-amd64.zip
!unzip ngrok-stable-linux-amd64.zip

LOG_DIR = './log'  # '"./drive/My Drive/log/"' for Drive location
get_ipython().system_raw(
    'tensorboard --logdir {} --host 0.0.0.0 --port 6006 &'
    .format(LOG_DIR)
)

import time

Kulikovpavel

# http://pytorch.org/
from os.path import exists
from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag
platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())
cuda_output = !ldconfig -p|grep cudart.so|sed -e 's/.*\.\([0-9]*\)\.\([0-9]*\)$/cu\1\2/'
accelerator = cuda_output[0] if exists('/dev/nvidia0') else 'cpu'
!pip install -q http://download.pytorch.org/whl/{accelerator}/torch-0.4.1-{platform}-linux_x86_64.whl torchvision

Kulikovpavel

*Programming. The metaphysics way.*, exercises for
==================================================

Class 1-alpha
-------------

We covered lambda calculus, some common operators, defining functions, using `.hs` files and GHCi and just barely scratched data types (GADTs; the `data Bool = False | True` part).

So here are some refreshers and clarifications for that, as well as something to go a bit further on with.