By now, you've used an example repository to review what an employer expects to read in a project notebook and README. In this Lab, you're asked to use the checklist to review your own project material. Use the lists below to review that each expected section is in your workbook and is able to answer the expected questions.
You will be able to:
- Review the minimum expectations of written data science project
- List what enhances a project from the minimum to the exceptional
- Compare the impact of rewritten sections to their original content
- Is the title desciptive?
- How much time will this solution save?
- Who will this solution help?
- What need does this analysis address?
- How well does the metric or target variable directly relate to the real world problem?
- Where does the data come from?
- What do the variables mean in actual language?
- What is the target varialbe?
- What is the range or scale of each variable?
- Who is in the sample or how was the data colllected?
- What elements of the data will or will not address the business question?
- Are there any issues in term of data permissions, copy right, ethical issues, or confidential information?
- Are there any interesting aspects or anomolies in the data such as outliers or missing data?
- What additional data would be really helpful in your analysis?
- Can someone else replicate your entire data preparation process?
- If you created the data through scraping or an API, can someone repeat that process?
- In what form is the data stored?
- There should be code that can take the raw data and get it ready for analysis, can be run again
- Is the code in pipeline form?
- Is all the preprocessing code in the notebook or is it in separate py files?
- Is the information you are including absolutley relelvant?
- Is your final model specified in an equation or pseudocode, and not just specified in code?
- When you describe the parameter or ceoficients, do you describe it in real terms?
- Have you examined any problems with the data that might be impacting the quallity of your analysi or model?
- What evaluation metrics did you use?
- Were there special considerations you made when choosing that evaluation metric?
- How does your model's metric compare to industry standards or what is already out there?
- Was cross vallidation included in your process and what concerns did that address?
- What are the limitations of interpreting your analysis?
- What next steps would you take in this analysis? What new data would you want to incorperate?
- How well does your analysis answer the actual business question and concern?
- What sort of impact would your results actually have?
- Detailed description of your business question
- A summary of your data science process, findings, and ideas future improvement
- At least one interesting visuallization from your analysis
- Repository navigation
- Links to the presentation slides, notebook, and other relevant documentatin
- Links to sources, such as the data, papers referenced, or other important materials
- Reproduction instructions
- How to contact you information
Congrats on making your Jupyter Notebook and README employer-ready!