CNN Architecture implementations in Tensorflow

[the-robot]

Your YT contents are awesome!! Especially those CNN architectures from scratch videos. I myself is trying to learn DL and your videos helped me understand the concept better when I was reading the academic papers.

Not very long ago, I started implementing some of the popular CNN architectures with Tensorflow 2.0 in my repo and I think it would be good to PR those to here so the rest can checkout both PyTorch and Tensorflow implementations.

I am not super good with Tensorflow, so if there's something that can be improved, feel free to give comments.

I have implemented

• AlexNet
• GoogLeNet / Inception V1
• LeNet5
• ResNet
• VGGNet

[aladdinpersson]

Hey that's awesome @the-robot! I just looked through it rather quickly but I think you could make different VGG types a little bit cleaner perhaps, take a look at my PyTorch implementation:

Machine-Learning-Collection/ML/Pytorch/CNN_architectures/pytorch_vgg_implementation.py

Lines 16 to 62 in 79f2e19

	VGG_types = {
	'VGG11': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
	'VGG13': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
	'VGG16': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
	'VGG19': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
	}
	class VGG_net(nn.Module):
	def __init__(self, in_channels=3, num_classes=1000):
	super(VGG_net, self).__init__()
	self.in_channels = in_channels
	self.conv_layers = self.create_conv_layers(VGG_types['VGG16'])
	self.fcs = nn.Sequential(
	nn.Linear(512*7*7, 4096),
	nn.ReLU(),
	nn.Dropout(p=0.5),
	nn.Linear(4096, 4096),
	nn.ReLU(),
	nn.Dropout(p=0.5),
	nn.Linear(4096, num_classes)
	)
	def forward(self, x):
	x = self.conv_layers(x)
	x = x.reshape(x.shape[0], -1)
	x = self.fcs(x)
	return x
	def create_conv_layers(self, architecture):
	layers = []
	in_channels = self.in_channels
	for x in architecture:
	if type(x) == int:
	out_channels = x
	layers += [nn.Conv2d(in_channels=in_channels,out_channels=out_channels,
	kernel_size=(3,3), stride=(1,1), padding=(1,1)),
	nn.BatchNorm2d(x),
	nn.ReLU()]
	in_channels = x
	elif x == 'M':
	layers += [nn.MaxPool2d(kernel_size=(2,2), stride=(2,2))]
	return nn.Sequential(*layers)

I think doing something like that with the VGG_types could make it a little more compact

[the-robot]

Hey that's awesome @the-robot! I just looked through it rather quickly but I think you could make different VGG types a little bit cleaner perhaps, take a look at my PyTorch implementation:

Machine-Learning-Collection/ML/Pytorch/CNN_architectures/pytorch_vgg_implementation.py

Lines 16 to 62 in 79f2e19

	VGG_types = {
	'VGG11': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
	'VGG13': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
	'VGG16': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
	'VGG19': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
	}
	class VGG_net(nn.Module):
	def __init__(self, in_channels=3, num_classes=1000):
	super(VGG_net, self).__init__()
	self.in_channels = in_channels
	self.conv_layers = self.create_conv_layers(VGG_types['VGG16'])
	self.fcs = nn.Sequential(
	nn.Linear(512*7*7, 4096),
	nn.ReLU(),
	nn.Dropout(p=0.5),
	nn.Linear(4096, 4096),
	nn.ReLU(),
	nn.Dropout(p=0.5),
	nn.Linear(4096, num_classes)
	)
	def forward(self, x):
	x = self.conv_layers(x)
	x = x.reshape(x.shape[0], -1)
	x = self.fcs(x)
	return x
	def create_conv_layers(self, architecture):
	layers = []
	in_channels = self.in_channels
	for x in architecture:
	if type(x) == int:
	out_channels = x
	layers += [nn.Conv2d(in_channels=in_channels,out_channels=out_channels,
	kernel_size=(3,3), stride=(1,1), padding=(1,1)),
	nn.BatchNorm2d(x),
	nn.ReLU()]
	in_channels = x
	elif x == 'M':
	layers += [nn.MaxPool2d(kernel_size=(2,2), stride=(2,2))]
	return nn.Sequential(*layers)

I think doing something like that with the VGG_types could make it a little more compact

Thanks for the suggestion. Initially I made it that way to group some layers together by following ConvNet config table from original paper. But storing different types in a dictionary would be much cleaner.

I also removed training part and put .summary() to show the model instead.

[aladdinpersson]

Added links in the readme, read through the implementations and looked really nice, liked the implementation of ResNet 👍