Essential session storage is required to sign in. Optional analytics remain off unless you allow them.
In a Jupyter notebook, take a real public dataset (a CSV of your choice — e.g. from Kaggle, data.gov, or a built-in seaborn dataset) and produce a clean, summarized analysis with pandas. Requirements: (1) LOAD the data with pd.read_csv, setting sensible options (parse_dates, dtype where needed); (2) INSPECT it — head/info/describe/shape and value_counts on a categorical column — and write a short note on what you found; (3) CLEAN it — handle missing values (dropna/fillna, stating your choice), remove duplicates, and fix at least one wrong dtype; (4) SUMMARIZE — use groupby to compute at least one 'metric per category' (e.g. average value per group) and derive one new column. State one thing the data shows.
Data loaded with sensible read_csv options; inspected with head/info/describe and value_counts, with a note on findings.
Missing values handled with a stated choice, duplicates removed, at least one dtype corrected (e.g. to_datetime).
A groupby aggregation (metric per category) plus one derived column; one clear observation stated.
Use this README structure to make the work reviewable.
DevPath fetches a read-only, size-capped snapshot. Submitted code is never executed.
# Load, clean, and summarize a real dataset ## Goal In a Jupyter notebook, take a real public dataset (a CSV of your choice — e.g. from Kaggle, data.gov, or a built-in seaborn dataset) and produce a clean, summarized analysis with pandas. Requirements: (1) LOAD the data with pd.read_csv, setting sensible options (parse_dates, dtype where needed); (2) INSPECT it — head/info/describe/shape and value_counts on a categorical column — and write a short note on what you found; (3) CLEAN it — handle missing values (dropna/fillna, stating your choice), remove duplicates, and fix at least one wrong dtype; (4) SUMMARIZE — use groupby to compute at least one 'metric per category' (e.g. average value per group) and derive one new column. State one thing the data shows. ## Acceptance criteria - [ ] Load & inspect: Data loaded with sensible read_csv options; inspected with head/info/describe and value_counts, with a note on findings. - [ ] Cleaning: Missing values handled with a stated choice, duplicates removed, at least one dtype corrected (e.g. to_datetime). - [ ] Summary: A groupby aggregation (metric per category) plus one derived column; one clear observation stated. ## Verification - [ ] Document installation and run commands. - [ ] Record automated checks and their exact commands. - [ ] Add representative output, screenshots, or a short demo where useful. - [ ] Confirm failure paths and known constraints. ## Decision log ### Decision title - Context: - Choice: - Alternatives considered: - Tradeoffs: ## Evidence - Link each rubric criterion to the file, test, or artifact that demonstrates it. ## Known limitations - List what is intentionally out of scope and what should be improved next.