This jupyter notebook based python project explores features of passengers who boarded the Titanic ship in 1912 to determine which features of the passengers best predict passengers who survived or did not survive using machine learning, we implement several conditional statements.
- titanic_survival_exploration.ipynb
- titanic_data.csv
- visuals.py
- titanic_survival_exploration.html
This section explains how to setup your system to run the codes found here.
- Download Anaconda setup for your operating system from here
- All installation instructions for Anaconda can be found here, but fo specifics, if you are on windows, follow the instructions here to setup anaconda and on linux here and for MAC OS from here
- If you have followed the instructions above, you should be able to run the jupyter notebook server from the terminal using $ jupyter-notebook and hit enter key. This will launch your web browser for the jupyter notebook from which you will load the file titanic_survival_exploration.ipynb
- Click the clone button at the top of this repository and take download .zip which will download this repository to your local computer usually to the Downloads folder of your computer. You will need Winrar to unzip the files to a dirctory called titanic_survival_exploration/ on your local system.
In order to be able to run the file titanic_survival_exploration.ipynb, you need to load a new kernel from the top right menu item on the browser tab that opened for jupyter. Click the Upload button and browse to the directory from which you extracted the file above to. Select titanic_survival_exploration.ipynb file and click open button to upload. Once uploaded to your jupyter notebook, click on the file name on the browser to open a new notebook for this project. Once you click, you should be able to see an interface with Graphs, Python codes, and questions with responses that were asked for this project. You can play around the codes following the instructions on the interfaces.
- From the notebook, the last prediction of 79.80% accuracy score does not meet the objective value of at least 80% for this project. The last code segment for question 4 does not carry the right conditions to meet this value.