Deep Learning Models -- A collection of various deep learning architectures, models, and tips for TensorFlow and PyTorch in Jupyter Notebooks.

In [1]:
%load_ext watermark
%watermark -a 'Sebastian Raschka' -v -p torch
Sebastian Raschka 

CPython 3.6.8
IPython 7.2.0

torch 1.0.0

Model Zoo -- CNN Gender Classifier (ResNet-18 Architecture, CelebA) with Data Parallelism

Network Architecture

The network in this notebook is an implementation of the ResNet-18 [1] architecture on the CelebA face dataset [2] to train a gender classifier.

References

  • [1] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778). (CVPR Link)

  • [2] Zhang, K., Tan, L., Li, Z., & Qiao, Y. (2016). Gender and smile classification using deep convolutional neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (pp. 34-38).

The following figure illustrates residual blocks with skip connections such that the input passed via the shortcut matches the dimensions of the main path's output, which allows the network to learn identity functions.

The ResNet-18 architecture actually uses residual blocks with skip connections such that the input passed via the shortcut matches is resized to dimensions of the main path's output. Such a residual block is illustrated below:

For a more detailed explanation see the other notebook, resnet-ex-1.ipynb.

Imports

In [2]:
import os
import time

import numpy as np
import pandas as pd

import torch
import torch.nn as nn
import torch.nn.functional as F

from torch.utils.data import Dataset
from torch.utils.data import DataLoader

from torchvision import datasets
from torchvision import transforms

import matplotlib.pyplot as plt
from PIL import Image


if torch.cuda.is_available():
    torch.backends.cudnn.deterministic = True

Settings

In [3]:
##########################
### SETTINGS
##########################

# Hyperparameters
RANDOM_SEED = 1
LEARNING_RATE = 0.001
NUM_EPOCHS = 10

# Architecture
NUM_FEATURES = 128*128
NUM_CLASSES = 2
BATCH_SIZE = 256*torch.cuda.device_count()
DEVICE = 'cuda:0' # default GPU device
GRAYSCALE = False

Dataset

Downloading the Dataset

Note that the ~200,000 CelebA face image dataset is relatively large (~1.3 Gb). The download link provided below was provided by the author on the official CelebA website at http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html.

1) Download and unzip the file img_align_celeba.zip, which contains the images in jpeg format.

2) Download the list_attr_celeba.txt file, which contains the class labels

3) Download the list_eval_partition.txt file, which contains training/validation/test partitioning info

Preparing the Dataset

In [4]:
df1 = pd.read_csv('list_attr_celeba.txt', sep="\s+", skiprows=1, usecols=['Male'])

# Make 0 (female) & 1 (male) labels instead of -1 & 1
df1.loc[df1['Male'] == -1, 'Male'] = 0

df1.head()
Out[4]:
Male
000001.jpg 0
000002.jpg 0
000003.jpg 1
000004.jpg 0
000005.jpg 0
In [5]:
df2 = pd.read_csv('list_eval_partition.txt', sep="\s+", skiprows=0, header=None)
df2.columns = ['Filename', 'Partition']
df2 = df2.set_index('Filename')

df2.head()
Out[5]:
Partition
Filename
000001.jpg 0
000002.jpg 0
000003.jpg 0
000004.jpg 0
000005.jpg 0
In [6]:
df3 = df1.merge(df2, left_index=True, right_index=True)
df3.head()
Out[6]:
Male Partition
000001.jpg 0 0
000002.jpg 0 0
000003.jpg 1 0
000004.jpg 0 0
000005.jpg 0 0
In [7]:
df3.to_csv('celeba-gender-partitions.csv')
df4 = pd.read_csv('celeba-gender-partitions.csv', index_col=0)
df4.head()
Out[7]:
Male Partition
000001.jpg 0 0
000002.jpg 0 0
000003.jpg 1 0
000004.jpg 0 0
000005.jpg 0 0
In [8]:
df4.loc[df4['Partition'] == 0].to_csv('celeba-gender-train.csv')
df4.loc[df4['Partition'] == 1].to_csv('celeba-gender-valid.csv')
df4.loc[df4['Partition'] == 2].to_csv('celeba-gender-test.csv')
In [9]:
img = Image.open('img_align_celeba/000001.jpg')
print(np.asarray(img, dtype=np.uint8).shape)
plt.imshow(img);
(218, 178, 3)

Implementing a Custom DataLoader Class

In [10]:
class CelebaDataset(Dataset):
    """Custom Dataset for loading CelebA face images"""

    def __init__(self, csv_path, img_dir, transform=None):
    
        df = pd.read_csv(csv_path, index_col=0)
        self.img_dir = img_dir
        self.csv_path = csv_path
        self.img_names = df.index.values
        self.y = df['Male'].values
        self.transform = transform

    def __getitem__(self, index):
        img = Image.open(os.path.join(self.img_dir,
                                      self.img_names[index]))
        
        if self.transform is not None:
            img = self.transform(img)
        
        label = self.y[index]
        return img, label

    def __len__(self):
        return self.y.shape[0]
In [11]:
# Note that transforms.ToTensor()
# already divides pixels by 255. internally

custom_transform = transforms.Compose([transforms.CenterCrop((178, 178)),
                                       transforms.Resize((128, 128)),
                                       #transforms.Grayscale(),                                       
                                       #transforms.Lambda(lambda x: x/255.),
                                       transforms.ToTensor()])

train_dataset = CelebaDataset(csv_path='celeba-gender-train.csv',
                              img_dir='img_align_celeba/',
                              transform=custom_transform)

valid_dataset = CelebaDataset(csv_path='celeba-gender-valid.csv',
                              img_dir='img_align_celeba/',
                              transform=custom_transform)

test_dataset = CelebaDataset(csv_path='celeba-gender-test.csv',
                             img_dir='img_align_celeba/',
                             transform=custom_transform)


train_loader = DataLoader(dataset=train_dataset,
                          batch_size=BATCH_SIZE,
                          shuffle=True,
                          num_workers=4)

valid_loader = DataLoader(dataset=valid_dataset,
                          batch_size=BATCH_SIZE,
                          shuffle=False,
                          num_workers=4)

test_loader = DataLoader(dataset=test_dataset,
                         batch_size=BATCH_SIZE,
                         shuffle=False,
                         num_workers=4)
In [12]:
torch.manual_seed(0)

for epoch in range(2):

    for batch_idx, (x, y) in enumerate(train_loader):
        
        print('Epoch:', epoch+1, end='')
        print(' | Batch index:', batch_idx, end='')
        print(' | Batch size:', y.size()[0])
        
        x = x.to(DEVICE)
        y = y.to(DEVICE)
        time.sleep(1)
        break
Epoch: 1 | Batch index: 0 | Batch size: 1024
Epoch: 2 | Batch index: 0 | Batch size: 1024

Model

The following code cell that implements the ResNet-34 architecture is a derivative of the code provided at https://pytorch.org/docs/0.4.0/_modules/torchvision/models/resnet.html.

In [13]:
##########################
### MODEL
##########################


def conv3x3(in_planes, out_planes, stride=1):
    """3x3 convolution with padding"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
                     padding=1, bias=False)


class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(BasicBlock, self).__init__()
        self.conv1 = conv3x3(inplanes, planes, stride)
        self.bn1 = nn.BatchNorm2d(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)
        self.bn2 = nn.BatchNorm2d(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        out += residual
        out = self.relu(out)

        return out




class ResNet(nn.Module):

    def __init__(self, block, layers, num_classes, grayscale):
        self.inplanes = 64
        if grayscale:
            in_dim = 1
        else:
            in_dim = 3
        super(ResNet, self).__init__()
        self.conv1 = nn.Conv2d(in_dim, 64, kernel_size=7, stride=2, padding=3,
                               bias=False)
        self.bn1 = nn.BatchNorm2d(64)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, layers[0])
        self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
        self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
        self.avgpool = nn.AvgPool2d(7, stride=1, padding=2)
        self.fc = nn.Linear(2048 * block.expansion, num_classes)

        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
                m.weight.data.normal_(0, (2. / n)**.5)
            elif isinstance(m, nn.BatchNorm2d):
                m.weight.data.fill_(1)
                m.bias.data.zero_()

    def _make_layer(self, block, planes, blocks, stride=1):
        downsample = None
        if stride != 1 or self.inplanes != planes * block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(self.inplanes, planes * block.expansion,
                          kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(planes * block.expansion),
            )

        layers = []
        layers.append(block(self.inplanes, planes, stride, downsample))
        self.inplanes = planes * block.expansion
        for i in range(1, blocks):
            layers.append(block(self.inplanes, planes))

        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avgpool(x)
        x = x.view(x.size(0), -1)
        logits = self.fc(x)
        probas = F.softmax(logits, dim=1)
        return logits, probas



def resnet18(num_classes):
    """Constructs a ResNet-18 model."""
    model = ResNet(block=BasicBlock, 
                   layers=[2, 2, 2, 2],
                   num_classes=NUM_CLASSES,
                   grayscale=GRAYSCALE)
    return model
In [14]:
torch.manual_seed(RANDOM_SEED)

##########################
### COST AND OPTIMIZER
##########################



model = resnet18(NUM_CLASSES)


#### DATA PARALLEL START ####
if torch.cuda.device_count() > 1:
    print("Using", torch.cuda.device_count(), "GPUs")
    model = nn.DataParallel(model)
#### DATA PARALLEL END ####


model.to(DEVICE)



cost_fn = torch.nn.CrossEntropyLoss()  
optimizer = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE)  
Using 4 GPUs

Training

In [15]:
def compute_accuracy(model, data_loader, device):
    correct_pred, num_examples = 0, 0
    for i, (features, targets) in enumerate(data_loader):
            
        features = features.to(device)
        targets = targets.to(device)

        logits, probas = model(features)
        _, predicted_labels = torch.max(probas, 1)
        num_examples += targets.size(0)
        correct_pred += (predicted_labels == targets).sum()
    return correct_pred.float()/num_examples * 100
    

start_time = time.time()
for epoch in range(NUM_EPOCHS):
    
    model.train()
    for batch_idx, (features, targets) in enumerate(train_loader):
        
        features = features.to(DEVICE)
        targets = targets.to(DEVICE)
            
        ### FORWARD AND BACK PROP
        logits, probas = model(features)
        cost = cost_fn(logits, targets)
        optimizer.zero_grad()
        
        cost.backward()
        
        ### UPDATE MODEL PARAMETERS
        optimizer.step()
        
        ### LOGGING
        if not batch_idx % 50:
            print ('Epoch: %03d/%03d | Batch %04d/%04d | Cost: %.4f' 
                   %(epoch+1, NUM_EPOCHS, batch_idx, 
                     len(train_loader), cost))

        

    model.eval()
    with torch.set_grad_enabled(False): # save memory during inference
        print('Epoch: %03d/%03d | Train: %.3f%% | Valid: %.3f%%' % (
              epoch+1, NUM_EPOCHS, 
              compute_accuracy(model, train_loader, device=DEVICE),
              compute_accuracy(model, valid_loader, device=DEVICE)))
        
    print('Time elapsed: %.2f min' % ((time.time() - start_time)/60))
    
print('Total Training Time: %.2f min' % ((time.time() - start_time)/60))
Epoch: 001/010 | Batch 0000/0159 | Cost: 0.6782
Epoch: 001/010 | Batch 0050/0159 | Cost: 0.1445
Epoch: 001/010 | Batch 0100/0159 | Cost: 0.1169
Epoch: 001/010 | Batch 0150/0159 | Cost: 0.0913
Epoch: 001/010 | Train: 93.687% | Valid: 94.101%
Time elapsed: 3.83 min
Epoch: 002/010 | Batch 0000/0159 | Cost: 0.0851
Epoch: 002/010 | Batch 0050/0159 | Cost: 0.0910
Epoch: 002/010 | Batch 0100/0159 | Cost: 0.0736
Epoch: 002/010 | Batch 0150/0159 | Cost: 0.0946
Epoch: 002/010 | Train: 96.940% | Valid: 97.025%
Time elapsed: 7.60 min
Epoch: 003/010 | Batch 0000/0159 | Cost: 0.0587
Epoch: 003/010 | Batch 0050/0159 | Cost: 0.0506
Epoch: 003/010 | Batch 0100/0159 | Cost: 0.0613
Epoch: 003/010 | Batch 0150/0159 | Cost: 0.0495
Epoch: 003/010 | Train: 98.260% | Valid: 97.896%
Time elapsed: 11.39 min
Epoch: 004/010 | Batch 0000/0159 | Cost: 0.0387
Epoch: 004/010 | Batch 0050/0159 | Cost: 0.0413
Epoch: 004/010 | Batch 0100/0159 | Cost: 0.0462
Epoch: 004/010 | Batch 0150/0159 | Cost: 0.0366
Epoch: 004/010 | Train: 98.561% | Valid: 97.705%
Time elapsed: 15.21 min
Epoch: 005/010 | Batch 0000/0159 | Cost: 0.0323
Epoch: 005/010 | Batch 0050/0159 | Cost: 0.0431
Epoch: 005/010 | Batch 0100/0159 | Cost: 0.0433
Epoch: 005/010 | Batch 0150/0159 | Cost: 0.0263
Epoch: 005/010 | Train: 98.692% | Valid: 97.715%
Time elapsed: 18.99 min
Epoch: 006/010 | Batch 0000/0159 | Cost: 0.0285
Epoch: 006/010 | Batch 0050/0159 | Cost: 0.0280
Epoch: 006/010 | Batch 0100/0159 | Cost: 0.0302
Epoch: 006/010 | Batch 0150/0159 | Cost: 0.0451
Epoch: 006/010 | Train: 98.880% | Valid: 97.730%
Time elapsed: 22.76 min
Epoch: 007/010 | Batch 0000/0159 | Cost: 0.0307
Epoch: 007/010 | Batch 0050/0159 | Cost: 0.0257
Epoch: 007/010 | Batch 0100/0159 | Cost: 0.0247
Epoch: 007/010 | Batch 0150/0159 | Cost: 0.0227
Epoch: 007/010 | Train: 99.276% | Valid: 97.966%
Time elapsed: 26.55 min
Epoch: 008/010 | Batch 0000/0159 | Cost: 0.0142
Epoch: 008/010 | Batch 0050/0159 | Cost: 0.0185
Epoch: 008/010 | Batch 0100/0159 | Cost: 0.0092
Epoch: 008/010 | Batch 0150/0159 | Cost: 0.0345
Epoch: 008/010 | Train: 99.536% | Valid: 97.972%
Time elapsed: 30.36 min
Epoch: 009/010 | Batch 0000/0159 | Cost: 0.0130
Epoch: 009/010 | Batch 0050/0159 | Cost: 0.0160
Epoch: 009/010 | Batch 0100/0159 | Cost: 0.0112
Epoch: 009/010 | Batch 0150/0159 | Cost: 0.0235
Epoch: 009/010 | Train: 99.211% | Valid: 97.926%
Time elapsed: 34.16 min
Epoch: 010/010 | Batch 0000/0159 | Cost: 0.0049
Epoch: 010/010 | Batch 0050/0159 | Cost: 0.0135
Epoch: 010/010 | Batch 0100/0159 | Cost: 0.0225
Epoch: 010/010 | Batch 0150/0159 | Cost: 0.0236
Epoch: 010/010 | Train: 99.520% | Valid: 97.972%
Time elapsed: 37.94 min
Total Training Time: 37.94 min

Evaluation

In [16]:
with torch.set_grad_enabled(False): # save memory during inference
    print('Test accuracy: %.2f%%' % (compute_accuracy(model, test_loader, device=DEVICE)))
Test accuracy: 97.38%
In [17]:
for batch_idx, (features, targets) in enumerate(test_loader):

    features = features
    targets = targets
    break
    
plt.imshow(np.transpose(features[0], (1, 2, 0)))
Out[17]:
<matplotlib.image.AxesImage at 0x7f9528f29da0>
In [18]:
model.eval()
logits, probas = model(features.to(DEVICE)[0, None])
print('Probability Female %.2f%%' % (probas[0][0]*100))
Probability Female 100.00%
In [19]:
%watermark -iv
numpy       1.15.4
pandas      0.23.4
torch       1.0.0
PIL.Image   5.3.0