Crack detectioin Classifier using RestNet9 Architect:

In this project, the Crack dataset will be used to identify the class of a given image whether it is an image with a crack or it does not have a crack. The used Neural network is a RestNet9 which is abriviation for Residual Block Network.

!pip install jovian --upgrade --quiet
project_name = 'restnet-crack-detection-project'

Import the dataset:

The dataset will be imported from the amazonaws repository. To do this, the following libraries must be imported

import os   # importing operating system library
from torchvision.datasets.utils import download_url # to download the data from a given url
from zipfile import ZipFile   # to extract zip files.

# Dowload the dataset
dataset_url = 'https://md-datasets-cache-zipfiles-prod.s3.eu-west-1.amazonaws.com/5y9wdsg2zt-2.zip'
download_url(dataset_url, '.');  # '.' to tell where to extract the folder, here in the current directory

Downloading https://md-datasets-cache-zipfiles-prod.s3.eu-west-1.amazonaws.com/5y9wdsg2zt-2.zip to ./5y9wdsg2zt-2.zip

HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))