A CNN-based deep learning model from ImageNet was utilized to extract image features. These were stored in a CSV file. Using these features, an SVM model was trained on 70% of the data to classify cat and dog images, then tested on the remaining 30%.
We used a DL model based on CNN which is preformed on ImageNet (taken from the site https://keras.io/api/applications/) to extract the characteristics of the images (feature Extract) and store these vectors in a CSV file. • using the extracted characteristics (the CSV file), form (Train) an SVM model for Classify the images as cat or dog on 70% of the dataset, and test it on 30% of the dataset.
you can download this data set from this link: https://www.kaggle.com/c/dogs-vs-cats/data.
the code in the file Tp_Cnn-Svm.ipynb the Training and resaults report in the file report.pdf.