This project aims to demonstrate object recognition on the CIFAR-10 dataset using the ResNet50 deep learning model. The CIFAR-10 dataset consists of 60,000 32x32 color images across 10 different classes, with 6,000 images per class. The ResNet50 model is a convolutional neural network architecture that has achieved state-of-the-art performance on various image classification tasks.
Before running the code, make sure you have the following dependencies installed:
-
Python (version >= 3.6)
-
TensorFlow (version >= 2.0)
-
Keras (version >= 2.4)
-
NumPy
-
Matplotlib
The CIFAR-10 dataset is not included in this repository. To download the dataset, please visit this website: Kaggle
Once downloaded, extract the dataset and place it in the data directory.
After training completes, the script will display the training and validation accuracy over each epoch. Additionally, it will save the model's performance metrics, such as accuracy and loss, in a CSV file named resnet50_cifar10_metrics.csv.
The accuracy also increases to 71% from 31% by using ResNet50.