0 ratings0% found this document useful (0 votes)
28 views57 pagesCNN Layer Visualization
This document discusses loading and visualizing a ResNet50 CNN model trained on ImageNet and applied to the Fashion MNIST dataset. It loads the pretrained ResNet50 model without the top layers, loads the Fashion MNIST training and test data, and summarizes the model architecture. The goal is to use the pretrained weights to classify images in the Fashion MNIST dataset.
Uploaded by
dolly nCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
28 views57 pagesCNN Layer Visualization
This document discusses loading and visualizing a ResNet50 CNN model trained on ImageNet and applied to the Fashion MNIST dataset. It loads the pretrained ResNet50 model without the top layers, loads the Fashion MNIST training and test data, and summarizes the model architecture. The goal is to use the pretrained weights to classify images in the Fashion MNIST dataset.
Uploaded by
dolly nCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 57
Fashion Mnist - Resnet 50 - CNN Visualization
Import Libraries
import tensorflow as tf
import pandas as pd
import numpy as np
from keras.models import Model
import matplotlib.pyplot as plt
import seaborn as sns
import random
from sklearn.metrics import confusion_matrix, classification_report,
accuracy_score
# from utils import mnist_reader # Download from Fashion Mnist repo
import mnist_reader
print(len(tf.config.list_physical_devices('GPU'))>0)
True
Download and Load Data
# Download Data if needed
# !wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-
images-idx3-ubyte.gz
# !wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-
labels-idx1-ubyte.gz
# !wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-
images-idx3-ubyte.gz
# !wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-
labels-idx1-ubyte.gz
# Load Data
# Change Data folder respectively
data_folder = "./data"
train_images, train_labels = mnist_reader.load_mnist(data_folder, "train")
test_images, test_labels = mnist_reader.load_mnist(data_folder, "t10k")
train_images = train_images.reshape(-1, 28,28)
test_images = test_images.reshape(-1, 28,28)
Load ResNet 50 Model Trained on ImageNet Data
# Load the pre trained model
# Top is not included for us to customize
base_model = tf.keras.applications.ResNet50(
include_top=False,
weights='imagenet')
# If u want to freeze
# base_model.trainable = False
base_model.summary()
Model: "resnet50"
_____________________________________________________________________________
_____________________
Layer (type) Output Shape Param # Connected to
=============================================================================
=====================
input_1 (InputLayer) [(None, None, None, 0 []
3)]
conv1_pad (ZeroPadding2D) (None, None, None, 0 ['input_1[0]
[0]']
3)
conv1_conv (Conv2D) (None, None, None, 9472
['conv1_pad[0][0]']
64)
conv1_bn (BatchNormalization) (None, None, None, 256
['conv1_conv[0][0]']
64)
conv1_relu (Activation) (None, None, None, 0
['conv1_bn[0][0]']
64)
pool1_pad (ZeroPadding2D) (None, None, None, 0
['conv1_relu[0][0]']
64)
pool1_pool (MaxPooling2D) (None, None, None, 0
['pool1_pad[0][0]']
64)
conv2_block1_1_conv (Conv2D) (None, None, None, 4160
['pool1_pool[0][0]']
64)
conv2_block1_1_bn (BatchNormal (None, None, None, 256
['conv2_block1_1_conv[0][0]']
ization) 64)
conv2_block1_1_relu (Activatio (None, None, None, 0
['conv2_block1_1_bn[0][0]']
n) 64)
conv2_block1_2_conv (Conv2D) (None, None, None, 36928
['conv2_block1_1_relu[0][0]']
64)
conv2_block1_2_bn (BatchNormal (None, None, None, 256
['conv2_block1_2_conv[0][0]']
ization) 64)
conv2_block1_2_relu (Activatio (None, None, None, 0
['conv2_block1_2_bn[0][0]']
n) 64)
conv2_block1_0_conv (Conv2D) (None, None, None, 16640
['pool1_pool[0][0]']
256)
conv2_block1_3_conv (Conv2D) (None, None, None, 16640
['conv2_block1_2_relu[0][0]']
256)
conv2_block1_0_bn (BatchNormal (None, None, None, 1024
['conv2_block1_0_conv[0][0]']
ization) 256)
conv2_block1_3_bn (BatchNormal (None, None, None, 1024
['conv2_block1_3_conv[0][0]']
ization) 256)
conv2_block1_add (Add) (None, None, None, 0
['conv2_block1_0_bn[0][0]',
256)
'conv2_block1_3_bn[0][0]']
conv2_block1_out (Activation) (None, None, None, 0
['conv2_block1_add[0][0]']
256)
conv2_block2_1_conv (Conv2D) (None, None, None, 16448
['conv2_block1_out[0][0]']
64)
conv2_block2_1_bn (BatchNormal (None, None, None, 256
['conv2_block2_1_conv[0][0]']
ization) 64)
conv2_block2_1_relu (Activatio (None, None, None, 0
['conv2_block2_1_bn[0][0]']
n) 64)
conv2_block2_2_conv (Conv2D) (None, None, None, 36928
['conv2_block2_1_relu[0][0]']
64)
conv2_block2_2_bn (BatchNormal (None, None, None, 256
['conv2_block2_2_conv[0][0]']
ization) 64)
conv2_block2_2_relu (Activatio (None, None, None, 0
['conv2_block2_2_bn[0][0]']
n) 64)
conv2_block2_3_conv (Conv2D) (None, None, None, 16640
['conv2_block2_2_relu[0][0]']
256)
conv2_block2_3_bn (BatchNormal (None, None, None, 1024
['conv2_block2_3_conv[0][0]']
ization) 256)
conv2_block2_add (Add) (None, None, None, 0
['conv2_block1_out[0][0]',
256)
'conv2_block2_3_bn[0][0]']
conv2_block2_out (Activation) (None, None, None, 0
['conv2_block2_add[0][0]']
256)
conv2_block3_1_conv (Conv2D) (None, None, None, 16448
['conv2_block2_out[0][0]']
64)
conv2_block3_1_bn (BatchNormal (None, None, None, 256
['conv2_block3_1_conv[0][0]']
ization) 64)
conv2_block3_1_relu (Activatio (None, None, None, 0
['conv2_block3_1_bn[0][0]']
n) 64)
conv2_block3_2_conv (Conv2D) (None, None, None, 36928
['conv2_block3_1_relu[0][0]']
64)
conv2_block3_2_bn (BatchNormal (None, None, None, 256
['conv2_block3_2_conv[0][0]']
ization) 64)
conv2_block3_2_relu (Activatio (None, None, None, 0
['conv2_block3_2_bn[0][0]']
n) 64)
conv2_block3_3_conv (Conv2D) (None, None, None, 16640
['conv2_block3_2_relu[0][0]']
256)
conv2_block3_3_bn (BatchNormal (None, None, None, 1024
['conv2_block3_3_conv[0][0]']
ization) 256)
conv2_block3_add (Add) (None, None, None, 0
['conv2_block2_out[0][0]',
256)
'conv2_block3_3_bn[0][0]']
conv2_block3_out (Activation) (None, None, None, 0
['conv2_block3_add[0][0]']
256)
conv3_block1_1_conv (Conv2D) (None, None, None, 32896
['conv2_block3_out[0][0]']
128)
conv3_block1_1_bn (BatchNormal (None, None, None, 512
['conv3_block1_1_conv[0][0]']
ization) 128)
conv3_block1_1_relu (Activatio (None, None, None, 0
['conv3_block1_1_bn[0][0]']
n) 128)
conv3_block1_2_conv (Conv2D) (None, None, None, 147584
['conv3_block1_1_relu[0][0]']
128)
conv3_block1_2_bn (BatchNormal (None, None, None, 512
['conv3_block1_2_conv[0][0]']
ization) 128)
conv3_block1_2_relu (Activatio (None, None, None, 0
['conv3_block1_2_bn[0][0]']
n) 128)
conv3_block1_0_conv (Conv2D) (None, None, None, 131584
['conv2_block3_out[0][0]']
512)
conv3_block1_3_conv (Conv2D) (None, None, None, 66048
['conv3_block1_2_relu[0][0]']
512)
conv3_block1_0_bn (BatchNormal (None, None, None, 2048
['conv3_block1_0_conv[0][0]']
ization) 512)
conv3_block1_3_bn (BatchNormal (None, None, None, 2048
['conv3_block1_3_conv[0][0]']
ization) 512)
conv3_block1_add (Add) (None, None, None, 0
['conv3_block1_0_bn[0][0]',
512)
'conv3_block1_3_bn[0][0]']
conv3_block1_out (Activation) (None, None, None, 0
['conv3_block1_add[0][0]']
512)
conv3_block2_1_conv (Conv2D) (None, None, None, 65664
['conv3_block1_out[0][0]']
128)
conv3_block2_1_bn (BatchNormal (None, None, None, 512
['conv3_block2_1_conv[0][0]']
ization) 128)
conv3_block2_1_relu (Activatio (None, None, None, 0
['conv3_block2_1_bn[0][0]']
n) 128)
conv3_block2_2_conv (Conv2D) (None, None, None, 147584
['conv3_block2_1_relu[0][0]']
128)
conv3_block2_2_bn (BatchNormal (None, None, None, 512
['conv3_block2_2_conv[0][0]']
ization) 128)
conv3_block2_2_relu (Activatio (None, None, None, 0
['conv3_block2_2_bn[0][0]']
n) 128)
conv3_block2_3_conv (Conv2D) (None, None, None, 66048
['conv3_block2_2_relu[0][0]']
512)
conv3_block2_3_bn (BatchNormal (None, None, None, 2048
['conv3_block2_3_conv[0][0]']
ization) 512)
conv3_block2_add (Add) (None, None, None, 0
['conv3_block1_out[0][0]',
512)
'conv3_block2_3_bn[0][0]']
conv3_block2_out (Activation) (None, None, None, 0
['conv3_block2_add[0][0]']
512)
conv3_block3_1_conv (Conv2D) (None, None, None, 65664
['conv3_block2_out[0][0]']
128)
conv3_block3_1_bn (BatchNormal (None, None, None, 512
['conv3_block3_1_conv[0][0]']
ization) 128)
conv3_block3_1_relu (Activatio (None, None, None, 0
['conv3_block3_1_bn[0][0]']
n) 128)
conv3_block3_2_conv (Conv2D) (None, None, None, 147584
['conv3_block3_1_relu[0][0]']
128)
conv3_block3_2_bn (BatchNormal (None, None, None, 512
['conv3_block3_2_conv[0][0]']
ization) 128)
conv3_block3_2_relu (Activatio (None, None, None, 0
['conv3_block3_2_bn[0][0]']
n) 128)
conv3_block3_3_conv (Conv2D) (None, None, None, 66048
['conv3_block3_2_relu[0][0]']
512)
conv3_block3_3_bn (BatchNormal (None, None, None, 2048
['conv3_block3_3_conv[0][0]']
ization) 512)
conv3_block3_add (Add) (None, None, None, 0
['conv3_block2_out[0][0]',
512)
'conv3_block3_3_bn[0][0]']
conv3_block3_out (Activation) (None, None, None, 0
['conv3_block3_add[0][0]']
512)
conv3_block4_1_conv (Conv2D) (None, None, None, 65664
['conv3_block3_out[0][0]']
128)
conv3_block4_1_bn (BatchNormal (None, None, None, 512
['conv3_block4_1_conv[0][0]']
ization) 128)
conv3_block4_1_relu (Activatio (None, None, None, 0
['conv3_block4_1_bn[0][0]']
n) 128)
conv3_block4_2_conv (Conv2D) (None, None, None, 147584
['conv3_block4_1_relu[0][0]']
128)
conv3_block4_2_bn (BatchNormal (None, None, None, 512
['conv3_block4_2_conv[0][0]']
ization) 128)
conv3_block4_2_relu (Activatio (None, None, None, 0
['conv3_block4_2_bn[0][0]']
n) 128)
conv3_block4_3_conv (Conv2D) (None, None, None, 66048
['conv3_block4_2_relu[0][0]']
512)
conv3_block4_3_bn (BatchNormal (None, None, None, 2048
['conv3_block4_3_conv[0][0]']
ization) 512)
conv3_block4_add (Add) (None, None, None, 0
['conv3_block3_out[0][0]',
512)
'conv3_block4_3_bn[0][0]']
conv3_block4_out (Activation) (None, None, None, 0
['conv3_block4_add[0][0]']
512)
conv4_block1_1_conv (Conv2D) (None, None, None, 131328
['conv3_block4_out[0][0]']
256)
conv4_block1_1_bn (BatchNormal (None, None, None, 1024
['conv4_block1_1_conv[0][0]']
ization) 256)
conv4_block1_1_relu (Activatio (None, None, None, 0
['conv4_block1_1_bn[0][0]']
n) 256)
conv4_block1_2_conv (Conv2D) (None, None, None, 590080
['conv4_block1_1_relu[0][0]']
256)
conv4_block1_2_bn (BatchNormal (None, None, None, 1024
['conv4_block1_2_conv[0][0]']
ization) 256)
conv4_block1_2_relu (Activatio (None, None, None, 0
['conv4_block1_2_bn[0][0]']
n) 256)
conv4_block1_0_conv (Conv2D) (None, None, None, 525312
['conv3_block4_out[0][0]']
1024)
conv4_block1_3_conv (Conv2D) (None, None, None, 263168
['conv4_block1_2_relu[0][0]']
1024)
conv4_block1_0_bn (BatchNormal (None, None, None, 4096
['conv4_block1_0_conv[0][0]']
ization) 1024)
conv4_block1_3_bn (BatchNormal (None, None, None, 4096
['conv4_block1_3_conv[0][0]']
ization) 1024)
conv4_block1_add (Add) (None, None, None, 0
['conv4_block1_0_bn[0][0]',
1024)
'conv4_block1_3_bn[0][0]']
conv4_block1_out (Activation) (None, None, None, 0
['conv4_block1_add[0][0]']
1024)
conv4_block2_1_conv (Conv2D) (None, None, None, 262400
['conv4_block1_out[0][0]']
256)
conv4_block2_1_bn (BatchNormal (None, None, None, 1024
['conv4_block2_1_conv[0][0]']
ization) 256)
conv4_block2_1_relu (Activatio (None, None, None, 0
['conv4_block2_1_bn[0][0]']
n) 256)
conv4_block2_2_conv (Conv2D) (None, None, None, 590080
['conv4_block2_1_relu[0][0]']
256)
conv4_block2_2_bn (BatchNormal (None, None, None, 1024
['conv4_block2_2_conv[0][0]']
ization) 256)
conv4_block2_2_relu (Activatio (None, None, None, 0
['conv4_block2_2_bn[0][0]']
n) 256)
conv4_block2_3_conv (Conv2D) (None, None, None, 263168
['conv4_block2_2_relu[0][0]']
1024)
conv4_block2_3_bn (BatchNormal (None, None, None, 4096
['conv4_block2_3_conv[0][0]']
ization) 1024)
conv4_block2_add (Add) (None, None, None, 0
['conv4_block1_out[0][0]',
1024)
'conv4_block2_3_bn[0][0]']
conv4_block2_out (Activation) (None, None, None, 0
['conv4_block2_add[0][0]']
1024)
conv4_block3_1_conv (Conv2D) (None, None, None, 262400
['conv4_block2_out[0][0]']
256)
conv4_block3_1_bn (BatchNormal (None, None, None, 1024
['conv4_block3_1_conv[0][0]']
ization) 256)
conv4_block3_1_relu (Activatio (None, None, None, 0
['conv4_block3_1_bn[0][0]']
n) 256)
conv4_block3_2_conv (Conv2D) (None, None, None, 590080
['conv4_block3_1_relu[0][0]']
256)
conv4_block3_2_bn (BatchNormal (None, None, None, 1024
['conv4_block3_2_conv[0][0]']
ization) 256)
conv4_block3_2_relu (Activatio (None, None, None, 0
['conv4_block3_2_bn[0][0]']
n) 256)
conv4_block3_3_conv (Conv2D) (None, None, None, 263168
['conv4_block3_2_relu[0][0]']
1024)
conv4_block3_3_bn (BatchNormal (None, None, None, 4096
['conv4_block3_3_conv[0][0]']
ization) 1024)
conv4_block3_add (Add) (None, None, None, 0
['conv4_block2_out[0][0]',
1024)
'conv4_block3_3_bn[0][0]']
conv4_block3_out (Activation) (None, None, None, 0
['conv4_block3_add[0][0]']
1024)
conv4_block4_1_conv (Conv2D) (None, None, None, 262400
['conv4_block3_out[0][0]']
256)
conv4_block4_1_bn (BatchNormal (None, None, None, 1024
['conv4_block4_1_conv[0][0]']
ization) 256)
conv4_block4_1_relu (Activatio (None, None, None, 0
['conv4_block4_1_bn[0][0]']
n) 256)
conv4_block4_2_conv (Conv2D) (None, None, None, 590080
['conv4_block4_1_relu[0][0]']
256)
conv4_block4_2_bn (BatchNormal (None, None, None, 1024
['conv4_block4_2_conv[0][0]']
ization) 256)
conv4_block4_2_relu (Activatio (None, None, None, 0
['conv4_block4_2_bn[0][0]']
n) 256)
conv4_block4_3_conv (Conv2D) (None, None, None, 263168
['conv4_block4_2_relu[0][0]']
1024)
conv4_block4_3_bn (BatchNormal (None, None, None, 4096
['conv4_block4_3_conv[0][0]']
ization) 1024)
conv4_block4_add (Add) (None, None, None, 0
['conv4_block3_out[0][0]',
1024)
'conv4_block4_3_bn[0][0]']
conv4_block4_out (Activation) (None, None, None, 0
['conv4_block4_add[0][0]']
1024)
conv4_block5_1_conv (Conv2D) (None, None, None, 262400
['conv4_block4_out[0][0]']
256)
conv4_block5_1_bn (BatchNormal (None, None, None, 1024
['conv4_block5_1_conv[0][0]']
ization) 256)
conv4_block5_1_relu (Activatio (None, None, None, 0
['conv4_block5_1_bn[0][0]']
n) 256)
conv4_block5_2_conv (Conv2D) (None, None, None, 590080
['conv4_block5_1_relu[0][0]']
256)
conv4_block5_2_bn (BatchNormal (None, None, None, 1024
['conv4_block5_2_conv[0][0]']
ization) 256)
conv4_block5_2_relu (Activatio (None, None, None, 0
['conv4_block5_2_bn[0][0]']
n) 256)
conv4_block5_3_conv (Conv2D) (None, None, None, 263168
['conv4_block5_2_relu[0][0]']
1024)
conv4_block5_3_bn (BatchNormal (None, None, None, 4096
['conv4_block5_3_conv[0][0]']
ization) 1024)
conv4_block5_add (Add) (None, None, None, 0
['conv4_block4_out[0][0]',
1024)
'conv4_block5_3_bn[0][0]']
conv4_block5_out (Activation) (None, None, None, 0
['conv4_block5_add[0][0]']
1024)
conv4_block6_1_conv (Conv2D) (None, None, None, 262400
['conv4_block5_out[0][0]']
256)
conv4_block6_1_bn (BatchNormal (None, None, None, 1024
['conv4_block6_1_conv[0][0]']
ization) 256)
conv4_block6_1_relu (Activatio (None, None, None, 0
['conv4_block6_1_bn[0][0]']
n) 256)
conv4_block6_2_conv (Conv2D) (None, None, None, 590080
['conv4_block6_1_relu[0][0]']
256)
conv4_block6_2_bn (BatchNormal (None, None, None, 1024
['conv4_block6_2_conv[0][0]']
ization) 256)
conv4_block6_2_relu (Activatio (None, None, None, 0
['conv4_block6_2_bn[0][0]']
n) 256)
conv4_block6_3_conv (Conv2D) (None, None, None, 263168
['conv4_block6_2_relu[0][0]']
1024)
conv4_block6_3_bn (BatchNormal (None, None, None, 4096
['conv4_block6_3_conv[0][0]']
ization) 1024)
conv4_block6_add (Add) (None, None, None, 0
['conv4_block5_out[0][0]',
1024)
'conv4_block6_3_bn[0][0]']
conv4_block6_out (Activation) (None, None, None, 0
['conv4_block6_add[0][0]']
1024)
conv5_block1_1_conv (Conv2D) (None, None, None, 524800
['conv4_block6_out[0][0]']
512)
conv5_block1_1_bn (BatchNormal (None, None, None, 2048
['conv5_block1_1_conv[0][0]']
ization) 512)
conv5_block1_1_relu (Activatio (None, None, None, 0
['conv5_block1_1_bn[0][0]']
n) 512)
conv5_block1_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block1_1_relu[0][0]']
512)
conv5_block1_2_bn (BatchNormal (None, None, None, 2048
['conv5_block1_2_conv[0][0]']
ization) 512)
conv5_block1_2_relu (Activatio (None, None, None, 0
['conv5_block1_2_bn[0][0]']
n) 512)
conv5_block1_0_conv (Conv2D) (None, None, None, 2099200
['conv4_block6_out[0][0]']
2048)
conv5_block1_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block1_2_relu[0][0]']
2048)
conv5_block1_0_bn (BatchNormal (None, None, None, 8192
['conv5_block1_0_conv[0][0]']
ization) 2048)
conv5_block1_3_bn (BatchNormal (None, None, None, 8192
['conv5_block1_3_conv[0][0]']
ization) 2048)
conv5_block1_add (Add) (None, None, None, 0
['conv5_block1_0_bn[0][0]',
2048)
'conv5_block1_3_bn[0][0]']
conv5_block1_out (Activation) (None, None, None, 0
['conv5_block1_add[0][0]']
2048)
conv5_block2_1_conv (Conv2D) (None, None, None, 1049088
['conv5_block1_out[0][0]']
512)
conv5_block2_1_bn (BatchNormal (None, None, None, 2048
['conv5_block2_1_conv[0][0]']
ization) 512)
conv5_block2_1_relu (Activatio (None, None, None, 0
['conv5_block2_1_bn[0][0]']
n) 512)
conv5_block2_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block2_1_relu[0][0]']
512)
conv5_block2_2_bn (BatchNormal (None, None, None, 2048
['conv5_block2_2_conv[0][0]']
ization) 512)
conv5_block2_2_relu (Activatio (None, None, None, 0
['conv5_block2_2_bn[0][0]']
n) 512)
conv5_block2_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block2_2_relu[0][0]']
2048)
conv5_block2_3_bn (BatchNormal (None, None, None, 8192
['conv5_block2_3_conv[0][0]']
ization) 2048)
conv5_block2_add (Add) (None, None, None, 0
['conv5_block1_out[0][0]',
2048)
'conv5_block2_3_bn[0][0]']
conv5_block2_out (Activation) (None, None, None, 0
['conv5_block2_add[0][0]']
2048)
conv5_block3_1_conv (Conv2D) (None, None, None, 1049088
['conv5_block2_out[0][0]']
512)
conv5_block3_1_bn (BatchNormal (None, None, None, 2048
['conv5_block3_1_conv[0][0]']
ization) 512)
conv5_block3_1_relu (Activatio (None, None, None, 0
['conv5_block3_1_bn[0][0]']
n) 512)
conv5_block3_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block3_1_relu[0][0]']
512)
conv5_block3_2_bn (BatchNormal (None, None, None, 2048
['conv5_block3_2_conv[0][0]']
ization) 512)
conv5_block3_2_relu (Activatio (None, None, None, 0
['conv5_block3_2_bn[0][0]']
n) 512)
conv5_block3_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block3_2_relu[0][0]']
2048)
conv5_block3_3_bn (BatchNormal (None, None, None, 8192
['conv5_block3_3_conv[0][0]']
ization) 2048)
conv5_block3_add (Add) (None, None, None, 0
['conv5_block2_out[0][0]',
2048)
'conv5_block3_3_bn[0][0]']
conv5_block3_out (Activation) (None, None, None, 0
['conv5_block3_add[0][0]']
2048)
=============================================================================
=====================
Total params: 23,587,712
Trainable params: 23,534,592
Non-trainable params: 53,120
_____________________________________________________________________________
_____________________
layer = base_model.layers #Conv layers at 2,7,10
filters, biases = base_model.layers[10].get_weights()
print(layer[2].name, filters.shape)
conv1_conv (3, 3, 64, 64)
fig1=plt.figure(figsize=(8, 12))
columns = 8
rows = 8
n_filters = columns * rows
for i in range(1, n_filters +1):
f = filters[:, :, :, i-1]
fig1 =plt.subplot(rows, columns, i)
fig1.set_xticks([]) #Turn off axis
fig1.set_yticks([])
plt.imshow(f[:, :, 0], cmap='gray') #Show only the filters from 0th
channel (R)
#ix += 1
plt.show()
conv_layer_index = [2, 7, 10] #TO define a shorter model
outputs = [base_model.layers[i].output for i in conv_layer_index]
model_short = Model(inputs=base_model.inputs, outputs=outputs)
print(model_short.summary())
Model: "model_3"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
input_1 (InputLayer) [(None, None, None, 3)] 0
conv1_pad (ZeroPadding2D) (None, None, None, 3) 0
conv1_conv (Conv2D) (None, None, None, 64) 9472
conv1_bn (BatchNormalizatio (None, None, None, 64) 256
n)
conv1_relu (Activation) (None, None, None, 64) 0
pool1_pad (ZeroPadding2D) (None, None, None, 64) 0
pool1_pool (MaxPooling2D) (None, None, None, 64) 0
conv2_block1_1_conv (Conv2D (None, None, None, 64) 4160
)
conv2_block1_1_bn (BatchNor (None, None, None, 64) 256
malization)
conv2_block1_1_relu (Activa (None, None, None, 64) 0
tion)
conv2_block1_2_conv (Conv2D (None, None, None, 64) 36928
)
=================================================================
Total params: 51,072
Trainable params: 50,816
Non-trainable params: 256
_________________________________________________________________
None
# from PIL import Image
# img_pil = Image.fromarray(test_images[2]).resize((28,28))
# im1 = img_pil.save("test_image2.png")
from tensorflow.keras.utils import load_img, img_to_array
img = load_img('test_image2.png', target_size=(28, 28)) #VGG user 224 as
input
img
img = img_to_array(img)
img = np.expand_dims(img, axis=0)
feature_output = model_short.predict(img)
1/1 [==============================] - 0s 84ms/step
columns = 8
rows = 8
for ftr in feature_output:
#pos = 1
print('layer')
fig=plt.figure(figsize=(12, 12))
for i in range(1, columns*rows +1):
fig =plt.subplot(rows, columns, i)
fig.set_xticks([]) #Turn off axis
fig.set_yticks([])
plt.imshow(ftr[0, :, :, i-1], cmap='gray')
#pos += 1
plt.show()
layer
layer
layer
---
.
from tensorflow.keras.utils import load_img, img_to_array
import numpy as np
import matplotlib.pyplot as plt
def get_cnn_visualization(model, list_of_layer_nos):
print(model.summary())
conv_layer_index = list_of_layer_nos #TO define a shorter model
outputs = [model.layers[i].output for i in conv_layer_index]
model_short = Model(inputs=base_model.inputs, outputs=outputs)
print(model_short.summary())
img = load_img('test_image2.png', target_size=(28, 28)) #VGG user 224 as
input
img = img_to_array(img)
img = np.expand_dims(img, axis=0)
feature_output = model_short.predict(img)
columns = 8
rows = 8
for ftr in feature_output:
#pos = 1
print('Conv Layer')
fig=plt.figure(figsize=(12, 12))
for i in range(1, columns*rows +1):
fig =plt.subplot(rows, columns, i)
fig.set_xticks([]) #Turn off axis
fig.set_yticks([])
plt.imshow(ftr[0, :, :, i-1], cmap='gray')
#pos += 1
plt.show()
from tensorflow.keras.utils import load_img, img_to_array
img = load_img('test_image2.png', target_size=(28, 28))
img
import tensorflow as tf
from keras.models import Model
from tensorflow.keras.utils import load_img, img_to_array
base_model = tf.keras.applications.ResNet50(
include_top=False,
weights='imagenet')
get_cnn_visualization(base_model, [2,7,10])
Model: "resnet50"
_____________________________________________________________________________
_____________________
Layer (type) Output Shape Param # Connected to
=============================================================================
=====================
input_18 (InputLayer) [(None, None, None, 0 []
3)]
conv1_pad (ZeroPadding2D) (None, None, None, 0
['input_18[0][0]']
3)
conv1_conv (Conv2D) (None, None, None, 9472
['conv1_pad[0][0]']
64)
conv1_bn (BatchNormalization) (None, None, None, 256
['conv1_conv[0][0]']
64)
conv1_relu (Activation) (None, None, None, 0
['conv1_bn[0][0]']
64)
pool1_pad (ZeroPadding2D) (None, None, None, 0
['conv1_relu[0][0]']
64)
pool1_pool (MaxPooling2D) (None, None, None, 0
['pool1_pad[0][0]']
64)
conv2_block1_1_conv (Conv2D) (None, None, None, 4160
['pool1_pool[0][0]']
64)
conv2_block1_1_bn (BatchNormal (None, None, None, 256
['conv2_block1_1_conv[0][0]']
ization) 64)
conv2_block1_1_relu (Activatio (None, None, None, 0
['conv2_block1_1_bn[0][0]']
n) 64)
conv2_block1_2_conv (Conv2D) (None, None, None, 36928
['conv2_block1_1_relu[0][0]']
64)
conv2_block1_2_bn (BatchNormal (None, None, None, 256
['conv2_block1_2_conv[0][0]']
ization) 64)
conv2_block1_2_relu (Activatio (None, None, None, 0
['conv2_block1_2_bn[0][0]']
n) 64)
conv2_block1_0_conv (Conv2D) (None, None, None, 16640
['pool1_pool[0][0]']
256)
conv2_block1_3_conv (Conv2D) (None, None, None, 16640
['conv2_block1_2_relu[0][0]']
256)
conv2_block1_0_bn (BatchNormal (None, None, None, 1024
['conv2_block1_0_conv[0][0]']
ization) 256)
conv2_block1_3_bn (BatchNormal (None, None, None, 1024
['conv2_block1_3_conv[0][0]']
ization) 256)
conv2_block1_add (Add) (None, None, None, 0
['conv2_block1_0_bn[0][0]',
256)
'conv2_block1_3_bn[0][0]']
conv2_block1_out (Activation) (None, None, None, 0
['conv2_block1_add[0][0]']
256)
conv2_block2_1_conv (Conv2D) (None, None, None, 16448
['conv2_block1_out[0][0]']
64)
conv2_block2_1_bn (BatchNormal (None, None, None, 256
['conv2_block2_1_conv[0][0]']
ization) 64)
conv2_block2_1_relu (Activatio (None, None, None, 0
['conv2_block2_1_bn[0][0]']
n) 64)
conv2_block2_2_conv (Conv2D) (None, None, None, 36928
['conv2_block2_1_relu[0][0]']
64)
conv2_block2_2_bn (BatchNormal (None, None, None, 256
['conv2_block2_2_conv[0][0]']
ization) 64)
conv2_block2_2_relu (Activatio (None, None, None, 0
['conv2_block2_2_bn[0][0]']
n) 64)
conv2_block2_3_conv (Conv2D) (None, None, None, 16640
['conv2_block2_2_relu[0][0]']
256)
conv2_block2_3_bn (BatchNormal (None, None, None, 1024
['conv2_block2_3_conv[0][0]']
ization) 256)
conv2_block2_add (Add) (None, None, None, 0
['conv2_block1_out[0][0]',
256)
'conv2_block2_3_bn[0][0]']
conv2_block2_out (Activation) (None, None, None, 0
['conv2_block2_add[0][0]']
256)
conv2_block3_1_conv (Conv2D) (None, None, None, 16448
['conv2_block2_out[0][0]']
64)
conv2_block3_1_bn (BatchNormal (None, None, None, 256
['conv2_block3_1_conv[0][0]']
ization) 64)
conv2_block3_1_relu (Activatio (None, None, None, 0
['conv2_block3_1_bn[0][0]']
n) 64)
conv2_block3_2_conv (Conv2D) (None, None, None, 36928
['conv2_block3_1_relu[0][0]']
64)
conv2_block3_2_bn (BatchNormal (None, None, None, 256
['conv2_block3_2_conv[0][0]']
ization) 64)
conv2_block3_2_relu (Activatio (None, None, None, 0
['conv2_block3_2_bn[0][0]']
n) 64)
conv2_block3_3_conv (Conv2D) (None, None, None, 16640
['conv2_block3_2_relu[0][0]']
256)
conv2_block3_3_bn (BatchNormal (None, None, None, 1024
['conv2_block3_3_conv[0][0]']
ization) 256)
conv2_block3_add (Add) (None, None, None, 0
['conv2_block2_out[0][0]',
256)
'conv2_block3_3_bn[0][0]']
conv2_block3_out (Activation) (None, None, None, 0
['conv2_block3_add[0][0]']
256)
conv3_block1_1_conv (Conv2D) (None, None, None, 32896
['conv2_block3_out[0][0]']
128)
conv3_block1_1_bn (BatchNormal (None, None, None, 512
['conv3_block1_1_conv[0][0]']
ization) 128)
conv3_block1_1_relu (Activatio (None, None, None, 0
['conv3_block1_1_bn[0][0]']
n) 128)
conv3_block1_2_conv (Conv2D) (None, None, None, 147584
['conv3_block1_1_relu[0][0]']
128)
conv3_block1_2_bn (BatchNormal (None, None, None, 512
['conv3_block1_2_conv[0][0]']
ization) 128)
conv3_block1_2_relu (Activatio (None, None, None, 0
['conv3_block1_2_bn[0][0]']
n) 128)
conv3_block1_0_conv (Conv2D) (None, None, None, 131584
['conv2_block3_out[0][0]']
512)
conv3_block1_3_conv (Conv2D) (None, None, None, 66048
['conv3_block1_2_relu[0][0]']
512)
conv3_block1_0_bn (BatchNormal (None, None, None, 2048
['conv3_block1_0_conv[0][0]']
ization) 512)
conv3_block1_3_bn (BatchNormal (None, None, None, 2048
['conv3_block1_3_conv[0][0]']
ization) 512)
conv3_block1_add (Add) (None, None, None, 0
['conv3_block1_0_bn[0][0]',
512)
'conv3_block1_3_bn[0][0]']
conv3_block1_out (Activation) (None, None, None, 0
['conv3_block1_add[0][0]']
512)
conv3_block2_1_conv (Conv2D) (None, None, None, 65664
['conv3_block1_out[0][0]']
128)
conv3_block2_1_bn (BatchNormal (None, None, None, 512
['conv3_block2_1_conv[0][0]']
ization) 128)
conv3_block2_1_relu (Activatio (None, None, None, 0
['conv3_block2_1_bn[0][0]']
n) 128)
conv3_block2_2_conv (Conv2D) (None, None, None, 147584
['conv3_block2_1_relu[0][0]']
128)
conv3_block2_2_bn (BatchNormal (None, None, None, 512
['conv3_block2_2_conv[0][0]']
ization) 128)
conv3_block2_2_relu (Activatio (None, None, None, 0
['conv3_block2_2_bn[0][0]']
n) 128)
conv3_block2_3_conv (Conv2D) (None, None, None, 66048
['conv3_block2_2_relu[0][0]']
512)
conv3_block2_3_bn (BatchNormal (None, None, None, 2048
['conv3_block2_3_conv[0][0]']
ization) 512)
conv3_block2_add (Add) (None, None, None, 0
['conv3_block1_out[0][0]',
512)
'conv3_block2_3_bn[0][0]']
conv3_block2_out (Activation) (None, None, None, 0
['conv3_block2_add[0][0]']
512)
conv3_block3_1_conv (Conv2D) (None, None, None, 65664
['conv3_block2_out[0][0]']
128)
conv3_block3_1_bn (BatchNormal (None, None, None, 512
['conv3_block3_1_conv[0][0]']
ization) 128)
conv3_block3_1_relu (Activatio (None, None, None, 0
['conv3_block3_1_bn[0][0]']
n) 128)
conv3_block3_2_conv (Conv2D) (None, None, None, 147584
['conv3_block3_1_relu[0][0]']
128)
conv3_block3_2_bn (BatchNormal (None, None, None, 512
['conv3_block3_2_conv[0][0]']
ization) 128)
conv3_block3_2_relu (Activatio (None, None, None, 0
['conv3_block3_2_bn[0][0]']
n) 128)
conv3_block3_3_conv (Conv2D) (None, None, None, 66048
['conv3_block3_2_relu[0][0]']
512)
conv3_block3_3_bn (BatchNormal (None, None, None, 2048
['conv3_block3_3_conv[0][0]']
ization) 512)
conv3_block3_add (Add) (None, None, None, 0
['conv3_block2_out[0][0]',
512)
'conv3_block3_3_bn[0][0]']
conv3_block3_out (Activation) (None, None, None, 0
['conv3_block3_add[0][0]']
512)
conv3_block4_1_conv (Conv2D) (None, None, None, 65664
['conv3_block3_out[0][0]']
128)
conv3_block4_1_bn (BatchNormal (None, None, None, 512
['conv3_block4_1_conv[0][0]']
ization) 128)
conv3_block4_1_relu (Activatio (None, None, None, 0
['conv3_block4_1_bn[0][0]']
n) 128)
conv3_block4_2_conv (Conv2D) (None, None, None, 147584
['conv3_block4_1_relu[0][0]']
128)
conv3_block4_2_bn (BatchNormal (None, None, None, 512
['conv3_block4_2_conv[0][0]']
ization) 128)
conv3_block4_2_relu (Activatio (None, None, None, 0
['conv3_block4_2_bn[0][0]']
n) 128)
conv3_block4_3_conv (Conv2D) (None, None, None, 66048
['conv3_block4_2_relu[0][0]']
512)
conv3_block4_3_bn (BatchNormal (None, None, None, 2048
['conv3_block4_3_conv[0][0]']
ization) 512)
conv3_block4_add (Add) (None, None, None, 0
['conv3_block3_out[0][0]',
512)
'conv3_block4_3_bn[0][0]']
conv3_block4_out (Activation) (None, None, None, 0
['conv3_block4_add[0][0]']
512)
conv4_block1_1_conv (Conv2D) (None, None, None, 131328
['conv3_block4_out[0][0]']
256)
conv4_block1_1_bn (BatchNormal (None, None, None, 1024
['conv4_block1_1_conv[0][0]']
ization) 256)
conv4_block1_1_relu (Activatio (None, None, None, 0
['conv4_block1_1_bn[0][0]']
n) 256)
conv4_block1_2_conv (Conv2D) (None, None, None, 590080
['conv4_block1_1_relu[0][0]']
256)
conv4_block1_2_bn (BatchNormal (None, None, None, 1024
['conv4_block1_2_conv[0][0]']
ization) 256)
conv4_block1_2_relu (Activatio (None, None, None, 0
['conv4_block1_2_bn[0][0]']
n) 256)
conv4_block1_0_conv (Conv2D) (None, None, None, 525312
['conv3_block4_out[0][0]']
1024)
conv4_block1_3_conv (Conv2D) (None, None, None, 263168
['conv4_block1_2_relu[0][0]']
1024)
conv4_block1_0_bn (BatchNormal (None, None, None, 4096
['conv4_block1_0_conv[0][0]']
ization) 1024)
conv4_block1_3_bn (BatchNormal (None, None, None, 4096
['conv4_block1_3_conv[0][0]']
ization) 1024)
conv4_block1_add (Add) (None, None, None, 0
['conv4_block1_0_bn[0][0]',
1024)
'conv4_block1_3_bn[0][0]']
conv4_block1_out (Activation) (None, None, None, 0
['conv4_block1_add[0][0]']
1024)
conv4_block2_1_conv (Conv2D) (None, None, None, 262400
['conv4_block1_out[0][0]']
256)
conv4_block2_1_bn (BatchNormal (None, None, None, 1024
['conv4_block2_1_conv[0][0]']
ization) 256)
conv4_block2_1_relu (Activatio (None, None, None, 0
['conv4_block2_1_bn[0][0]']
n) 256)
conv4_block2_2_conv (Conv2D) (None, None, None, 590080
['conv4_block2_1_relu[0][0]']
256)
conv4_block2_2_bn (BatchNormal (None, None, None, 1024
['conv4_block2_2_conv[0][0]']
ization) 256)
conv4_block2_2_relu (Activatio (None, None, None, 0
['conv4_block2_2_bn[0][0]']
n) 256)
conv4_block2_3_conv (Conv2D) (None, None, None, 263168
['conv4_block2_2_relu[0][0]']
1024)
conv4_block2_3_bn (BatchNormal (None, None, None, 4096
['conv4_block2_3_conv[0][0]']
ization) 1024)
conv4_block2_add (Add) (None, None, None, 0
['conv4_block1_out[0][0]',
1024)
'conv4_block2_3_bn[0][0]']
conv4_block2_out (Activation) (None, None, None, 0
['conv4_block2_add[0][0]']
1024)
conv4_block3_1_conv (Conv2D) (None, None, None, 262400
['conv4_block2_out[0][0]']
256)
conv4_block3_1_bn (BatchNormal (None, None, None, 1024
['conv4_block3_1_conv[0][0]']
ization) 256)
conv4_block3_1_relu (Activatio (None, None, None, 0
['conv4_block3_1_bn[0][0]']
n) 256)
conv4_block3_2_conv (Conv2D) (None, None, None, 590080
['conv4_block3_1_relu[0][0]']
256)
conv4_block3_2_bn (BatchNormal (None, None, None, 1024
['conv4_block3_2_conv[0][0]']
ization) 256)
conv4_block3_2_relu (Activatio (None, None, None, 0
['conv4_block3_2_bn[0][0]']
n) 256)
conv4_block3_3_conv (Conv2D) (None, None, None, 263168
['conv4_block3_2_relu[0][0]']
1024)
conv4_block3_3_bn (BatchNormal (None, None, None, 4096
['conv4_block3_3_conv[0][0]']
ization) 1024)
conv4_block3_add (Add) (None, None, None, 0
['conv4_block2_out[0][0]',
1024)
'conv4_block3_3_bn[0][0]']
conv4_block3_out (Activation) (None, None, None, 0
['conv4_block3_add[0][0]']
1024)
conv4_block4_1_conv (Conv2D) (None, None, None, 262400
['conv4_block3_out[0][0]']
256)
conv4_block4_1_bn (BatchNormal (None, None, None, 1024
['conv4_block4_1_conv[0][0]']
ization) 256)
conv4_block4_1_relu (Activatio (None, None, None, 0
['conv4_block4_1_bn[0][0]']
n) 256)
conv4_block4_2_conv (Conv2D) (None, None, None, 590080
['conv4_block4_1_relu[0][0]']
256)
conv4_block4_2_bn (BatchNormal (None, None, None, 1024
['conv4_block4_2_conv[0][0]']
ization) 256)
conv4_block4_2_relu (Activatio (None, None, None, 0
['conv4_block4_2_bn[0][0]']
n) 256)
conv4_block4_3_conv (Conv2D) (None, None, None, 263168
['conv4_block4_2_relu[0][0]']
1024)
conv4_block4_3_bn (BatchNormal (None, None, None, 4096
['conv4_block4_3_conv[0][0]']
ization) 1024)
conv4_block4_add (Add) (None, None, None, 0
['conv4_block3_out[0][0]',
1024)
'conv4_block4_3_bn[0][0]']
conv4_block4_out (Activation) (None, None, None, 0
['conv4_block4_add[0][0]']
1024)
conv4_block5_1_conv (Conv2D) (None, None, None, 262400
['conv4_block4_out[0][0]']
256)
conv4_block5_1_bn (BatchNormal (None, None, None, 1024
['conv4_block5_1_conv[0][0]']
ization) 256)
conv4_block5_1_relu (Activatio (None, None, None, 0
['conv4_block5_1_bn[0][0]']
n) 256)
conv4_block5_2_conv (Conv2D) (None, None, None, 590080
['conv4_block5_1_relu[0][0]']
256)
conv4_block5_2_bn (BatchNormal (None, None, None, 1024
['conv4_block5_2_conv[0][0]']
ization) 256)
conv4_block5_2_relu (Activatio (None, None, None, 0
['conv4_block5_2_bn[0][0]']
n) 256)
conv4_block5_3_conv (Conv2D) (None, None, None, 263168
['conv4_block5_2_relu[0][0]']
1024)
conv4_block5_3_bn (BatchNormal (None, None, None, 4096
['conv4_block5_3_conv[0][0]']
ization) 1024)
conv4_block5_add (Add) (None, None, None, 0
['conv4_block4_out[0][0]',
1024)
'conv4_block5_3_bn[0][0]']
conv4_block5_out (Activation) (None, None, None, 0
['conv4_block5_add[0][0]']
1024)
conv4_block6_1_conv (Conv2D) (None, None, None, 262400
['conv4_block5_out[0][0]']
256)
conv4_block6_1_bn (BatchNormal (None, None, None, 1024
['conv4_block6_1_conv[0][0]']
ization) 256)
conv4_block6_1_relu (Activatio (None, None, None, 0
['conv4_block6_1_bn[0][0]']
n) 256)
conv4_block6_2_conv (Conv2D) (None, None, None, 590080
['conv4_block6_1_relu[0][0]']
256)
conv4_block6_2_bn (BatchNormal (None, None, None, 1024
['conv4_block6_2_conv[0][0]']
ization) 256)
conv4_block6_2_relu (Activatio (None, None, None, 0
['conv4_block6_2_bn[0][0]']
n) 256)
conv4_block6_3_conv (Conv2D) (None, None, None, 263168
['conv4_block6_2_relu[0][0]']
1024)
conv4_block6_3_bn (BatchNormal (None, None, None, 4096
['conv4_block6_3_conv[0][0]']
ization) 1024)
conv4_block6_add (Add) (None, None, None, 0
['conv4_block5_out[0][0]',
1024)
'conv4_block6_3_bn[0][0]']
conv4_block6_out (Activation) (None, None, None, 0
['conv4_block6_add[0][0]']
1024)
conv5_block1_1_conv (Conv2D) (None, None, None, 524800
['conv4_block6_out[0][0]']
512)
conv5_block1_1_bn (BatchNormal (None, None, None, 2048
['conv5_block1_1_conv[0][0]']
ization) 512)
conv5_block1_1_relu (Activatio (None, None, None, 0
['conv5_block1_1_bn[0][0]']
n) 512)
conv5_block1_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block1_1_relu[0][0]']
512)
conv5_block1_2_bn (BatchNormal (None, None, None, 2048
['conv5_block1_2_conv[0][0]']
ization) 512)
conv5_block1_2_relu (Activatio (None, None, None, 0
['conv5_block1_2_bn[0][0]']
n) 512)
conv5_block1_0_conv (Conv2D) (None, None, None, 2099200
['conv4_block6_out[0][0]']
2048)
conv5_block1_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block1_2_relu[0][0]']
2048)
conv5_block1_0_bn (BatchNormal (None, None, None, 8192
['conv5_block1_0_conv[0][0]']
ization) 2048)
conv5_block1_3_bn (BatchNormal (None, None, None, 8192
['conv5_block1_3_conv[0][0]']
ization) 2048)
conv5_block1_add (Add) (None, None, None, 0
['conv5_block1_0_bn[0][0]',
2048)
'conv5_block1_3_bn[0][0]']
conv5_block1_out (Activation) (None, None, None, 0
['conv5_block1_add[0][0]']
2048)
conv5_block2_1_conv (Conv2D) (None, None, None, 1049088
['conv5_block1_out[0][0]']
512)
conv5_block2_1_bn (BatchNormal (None, None, None, 2048
['conv5_block2_1_conv[0][0]']
ization) 512)
conv5_block2_1_relu (Activatio (None, None, None, 0
['conv5_block2_1_bn[0][0]']
n) 512)
conv5_block2_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block2_1_relu[0][0]']
512)
conv5_block2_2_bn (BatchNormal (None, None, None, 2048
['conv5_block2_2_conv[0][0]']
ization) 512)
conv5_block2_2_relu (Activatio (None, None, None, 0
['conv5_block2_2_bn[0][0]']
n) 512)
conv5_block2_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block2_2_relu[0][0]']
2048)
conv5_block2_3_bn (BatchNormal (None, None, None, 8192
['conv5_block2_3_conv[0][0]']
ization) 2048)
conv5_block2_add (Add) (None, None, None, 0
['conv5_block1_out[0][0]',
2048)
'conv5_block2_3_bn[0][0]']
conv5_block2_out (Activation) (None, None, None, 0
['conv5_block2_add[0][0]']
2048)
conv5_block3_1_conv (Conv2D) (None, None, None, 1049088
['conv5_block2_out[0][0]']
512)
conv5_block3_1_bn (BatchNormal (None, None, None, 2048
['conv5_block3_1_conv[0][0]']
ization) 512)
conv5_block3_1_relu (Activatio (None, None, None, 0
['conv5_block3_1_bn[0][0]']
n) 512)
conv5_block3_2_conv (Conv2D) (None, None, None, 2359808
['conv5_block3_1_relu[0][0]']
512)
conv5_block3_2_bn (BatchNormal (None, None, None, 2048
['conv5_block3_2_conv[0][0]']
ization) 512)
conv5_block3_2_relu (Activatio (None, None, None, 0
['conv5_block3_2_bn[0][0]']
n) 512)
conv5_block3_3_conv (Conv2D) (None, None, None, 1050624
['conv5_block3_2_relu[0][0]']
2048)
conv5_block3_3_bn (BatchNormal (None, None, None, 8192
['conv5_block3_3_conv[0][0]']
ization) 2048)
conv5_block3_add (Add) (None, None, None, 0
['conv5_block2_out[0][0]',
2048)
'conv5_block3_3_bn[0][0]']
conv5_block3_out (Activation) (None, None, None, 0
['conv5_block3_add[0][0]']
2048)
=============================================================================
=====================
Total params: 23,587,712
Trainable params: 23,534,592
Non-trainable params: 53,120
_____________________________________________________________________________
_____________________
None
Model: "model_16"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
input_18 (InputLayer) [(None, None, None, 3)] 0
conv1_pad (ZeroPadding2D) (None, None, None, 3) 0
conv1_conv (Conv2D) (None, None, None, 64) 9472
conv1_bn (BatchNormalizatio (None, None, None, 64) 256
n)
conv1_relu (Activation) (None, None, None, 64) 0
pool1_pad (ZeroPadding2D) (None, None, None, 64) 0
pool1_pool (MaxPooling2D) (None, None, None, 64) 0
conv2_block1_1_conv (Conv2D (None, None, None, 64) 4160
)
conv2_block1_1_bn (BatchNor (None, None, None, 64) 256
malization)
conv2_block1_1_relu (Activa (None, None, None, 64) 0
tion)
conv2_block1_2_conv (Conv2D (None, None, None, 64) 36928
)
=================================================================
Total params: 51,072
Trainable params: 50,816
Non-trainable params: 256
_________________________________________________________________
None
1/1 [==============================] - 0s 269ms/step
Conv Layer
Conv Layer
Conv Layer
You might also like
- BEH41803. Using Google Colab To Train Image From Folder. Dog Vs Cat Step 1: Connect To Google DriveNo ratings yetBEH41803. Using Google Colab To Train Image From Folder. Dog Vs Cat Step 1: Connect To Google Drive7 pages
- Build - CIFAR - 10 - Dataset - Classifier - Ipynb - ColaboratoryNo ratings yetBuild - CIFAR - 10 - Dataset - Classifier - Ipynb - Colaboratory8 pages
- 3 Salazar Francisco Improving - Accuracy - Using - ConvolutionsNo ratings yet3 Salazar Francisco Improving - Accuracy - Using - Convolutions14 pages
- Chapter08 - Intro To DL For Computer VisionNo ratings yetChapter08 - Intro To DL For Computer Vision10 pages
- Transfer Learning Alexnet.ipynb - ColaboratoryNo ratings yetTransfer Learning Alexnet.ipynb - Colaboratory5 pages
- dog and cat img classification.ipynb - ColabNo ratings yetdog and cat img classification.ipynb - Colab5 pages
- Tomato-Leaf-Disease-Classification - Ipynb - ColaboratoryNo ratings yetTomato-Leaf-Disease-Classification - Ipynb - Colaboratory7 pages
- Kecerdasan Artifisial Dan Masyarakat - M5No ratings yetKecerdasan Artifisial Dan Masyarakat - M58 pages
- Deep Learning Using Python + Keras (Chapter 3) - ResNet - CodeProjectNo ratings yetDeep Learning Using Python + Keras (Chapter 3) - ResNet - CodeProject24 pages
- Developing Website Life Starter: Pushadapu Bhavani 20951A0522 Indu Priya Tentu 20951A0556 Abhishek 20951A0505No ratings yetDeveloping Website Life Starter: Pushadapu Bhavani 20951A0522 Indu Priya Tentu 20951A0556 Abhishek 20951A050552 pages
- AI in Logistics and Supply Chain ManagementNo ratings yetAI in Logistics and Supply Chain Management19 pages
- Van Der Vlist Et Al 2024 Big Ai Cloud Infrastructure Dependence and The Industrialisation of Artificial IntelligenceNo ratings yetVan Der Vlist Et Al 2024 Big Ai Cloud Infrastructure Dependence and The Industrialisation of Artificial Intelligence16 pages
- Machine Learning Protocol From Ultrasound Data For Monitoring Predicting and Sup100% (1)Machine Learning Protocol From Ultrasound Data For Monitoring Predicting and Sup5 pages
- Topic: Criminal Liability of Artificial Intelligence.: Ilma Khan Roll No: 39 B.A.LLB 2 Year Criminal LawNo ratings yetTopic: Criminal Liability of Artificial Intelligence.: Ilma Khan Roll No: 39 B.A.LLB 2 Year Criminal Law13 pages
- Defect Prediction: Using Machine Learning: Kirti Hegde, Consultant Trupti Songadwala, Senior Consultant DeloitteNo ratings yetDefect Prediction: Using Machine Learning: Kirti Hegde, Consultant Trupti Songadwala, Senior Consultant Deloitte13 pages
- Detection of Employee Stress Using Machine LearningNo ratings yetDetection of Employee Stress Using Machine Learning14 pages
- Computational Data Science: Advanced Programme inNo ratings yetComputational Data Science: Advanced Programme in15 pages
- Artificial Intelligence in Knowledge Management Overview and Trend100% (1)Artificial Intelligence in Knowledge Management Overview and Trend8 pages
- A Machine Learning Framework For Resource Allocation Assisted by Cloud ComputingNo ratings yetA Machine Learning Framework For Resource Allocation Assisted by Cloud Computing8 pages
- EEE604 Electrical Engineering Modeling UD PDFNo ratings yetEEE604 Electrical Engineering Modeling UD PDF4 pages