CS x476 - Fall 2021

Project 3: Scene Recognition with Deep Learning

Refer jupyter notebook for info on dataset

Setup:

1. Install Miniconda. (If you already have Miniconda installed, you can skip this step)
2. Create a conda environment using the appropriate terminal and command. ● On Windows , open the installed "Anaconda Powershell Prompt". ● On MacOS and Linux , you can open a terminal window. ● Modify and run the command in the terminal, replace the “” in the following command with your OS (Linux, Mac, Windows): conda env create -f cv_proj3_configs/cv_proj3_env_.yml
3. Check if the cv_proj3 environment has been created properly. ● Run: conda env list
4. Activate the conda environment. ● Run: conda activate cv_proj ● To deactivate it, run: conda deactivate
5. Install the project packages. ● Run: pip install -e. inside the repo folder. ● This is a good practice when setting up a new conda environment that may have pip requirements. It installs the repo as a package in the environment.
6. Open the jupyter notebook to work on the project. ● Run: jupyter notebook ./cv_proj3_code/proj3.ipynb