Using Densenet and PyTorch

In this project i used Densenet for image classification in PyTorch with custom sampling function for pytorch imbalanced-dataset-sampler

Final Project Notebook

1. Installation

Download Anaconda

Install Anaconda on your machine. Detailed instructions:

2. Create and Activate the Environment

Please go though this doc before you creating an environment. After that create a environment using following command

conda create --name deep-learning

Then activate the environment using following command

activate deep-learning

Git and version control

These instructions also assume you have git installed for working with Github from a terminal window, but if you do not, you can download that first with the command:

conda install git

Now, you can create a local version of the project

1. Clone the repository, and navigate to the downloaded folder. This may take a minute or two to clone due to the included image data.

git clone https://github.com/koushik-elite/Using-Densenet-and-PyTorch.git
cd Using-Densenet-and-PyTorch

2. Install PyTorch and torchvision; this should install the latest version of PyTorch.

   conda install pytorch torchvision -c pytorch

   conda install pytorch -c pytorch
   pip install torchvision

• Windows:

• Linux or Mac:

3. Install a few required pip packages, which are specified in the requirements text file (including OpenCV).

pip install -r requirements.txt

4. That's it!, Now run the project using following command, check you default browser and open "Using Densenet and PyTorch.ipynb" file

jupyter notebook

Approach for Image Classification

Curently iam working on t-Distributed Stochastic Neighbor Embedding

References:

kaggle-1-winning-approach-for-image-classification-challenge from Kumar Shridhar

t-Distributed Stochastic Neighbor Embedding from Laurens van der Maaten

transfer-learning-the-art-of-fine-tuning-a-pre-trained-model from DISHASHREE GUPTA