This repository contains two Jupyter Notebooks that serve as tutorials in Polish for performing image segmentation and image classification using PyTorch. The tutorials guide you through the process of loading datasets, defining and training models, and creating interactive interfaces with Gradio.
- Image Segmentation with DeepLabV3 on VOCSegmentation Dataset
- Image Classification with Custom CNN and ResNet-18 on Caltech101 Dataset
This notebook demonstrates how to use a pre-trained DeepLabV3 model for image segmentation on the VOCSegmentation dataset. The tutorial covers the following steps:
Importing necessary libraries Defining data transformations and loading the dataset Setting up the pre-trained DeepLabV3 model Training the model Saving the trained model Loading the trained model and performing image segmentation Creating an interactive interface with Gradio Colab Link: Open in Colab
This notebook demonstrates how to build and train a custom CNN and fine-tune a pre-trained ResNet-18 model for image classification on the Caltech101 dataset. The tutorial covers the following steps:
Importing necessary libraries Defining data transformations and loading the dataset Creating a custom CNN model Loading and modifying a pre-trained ResNet-18 model Defining loss functions and optimizers Training both models Saving the trained ResNet-18 model Loading the trained model and creating an interactive interface with Gradio Colab Link: Open in Colab
Both notebooks are written in Polish to provide a detailed tutorial experience for Polish-speaking users. They include explanations and code comments to help you understand the key concepts and steps involved in image segmentation and classification using PyTorch.