PyTorch_Image_Captioning

PyTorch Image Captioning CNN-RNN Model.

In this project, we create and train a neural network to automatically generate captions from images.

After using the Microsoft Common Objects in COntext (MS COCO) dataset to train the network, we test it on novel images!

The project is structured as a series of sequential Jupyter notebooks:

0_Dataset.ipynb

1_Preliminaries.ipynb

2_Training.ipynb

3_Inference.ipynb

The model.py file contains the CNN-RNN model architecture and methods.

All the information and steps needed to understand this project from beginning to end are contained within the 4 Jupyter notebooks.

You must have a GPU for this project.