This is a set of notebooks designed for basic introductory level training for all data scientists. The idea is that a data scientist should be able to express oneself using pandas easily, as without pandas, it would be difficult to wrangle data quickly and efficiently.
To get started with this repo, please open questions.ipynb, the notebook that contains a series of questions that stem from the famous Titanic dataset.
Once you have attempted all the questions, please refer to the solutions.ipynb file to see the answers.
As there are many categorical variables in this dataset, an altered version has been created and uploaded to Hugging Face: https://huggingface.co/datasets/sigtica/titantic_altered.csv
This demonstrates the power of one-hot encoding and interpreting the data. For example, are all variables with fewer than, say, 10 unique values, categorical? It is really open to interpretation.
Please see titanic_altered.ipynb for more details.
If you have any questions or spot any errors, please contact info@sigtica.com. Please refer to pandas documentation first: https://pandas.pydata.org/docs/.