> For the complete documentation index, see [llms.txt](https://opencampus.gitbook.io/opencampus-machine-learning-program/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://opencampus.gitbook.io/opencampus-machine-learning-program/courses/einfuehrung/week-02-import-and-visualization.md).
# Week 2 - Data Import and Visualization
**Estimated workload:** about 6.5 hours (session 1.75 h, videos with notes 1.7 h, tasks 3 h)
**Question of the week (from the operations manager):** *On which weekdays should we bake more, and how sure can we be?*
**Canvas focus:** block 3 (Audiences): no new canvas this week; your memo is for the operations manager in your week 1 canvas · **Skills:** `audience` (loaded by the agent when it writes for a reader)
### This week you will ...
* steer a coding agent with clear context and verify its results with a short checklist,
* select, filter, sort and summarise data with pandas,
* choose suitable chart types, show uncertainty with confidence intervals, and write an action title that states the conclusion.
### Until next week you should ...
* [x] **Watch these videos first** (≈ 22 minutes, you need them for the tasks): *Segmentation by one feature*, *Segmentation by multiple features*, *Grouped column charts*, *Matplotlib subplots*, *Colors, grids, & saving plots* and *Understanding confidence intervals* (all in the table below).
* [x] **Finish the weekday analysis** (≈ 1 hour): everyone works in their own notebook `week-02-weekdays.ipynb` in the main folder of their practice repository, next to `week-01-seasonal-bread.ipynb`. If the notebook from the session is on someone else’s laptop, or you missed the session, start it with this prompt:
```
Create the notebook week-02-weekdays.ipynb in the main folder of this repository, not in a step folder. Load umsatzdaten_gekuerzt.csv from the URL in AGENTS.md. Add up the turnover per day, then plot the average daily turnover per weekday (Monday to Sunday) with 95 % confidence intervals: one chart for all products together, and one subplot per product group. Explain each step.
```
Use one subplot per product group: the groups sell very different amounts, so in one combined chart the small groups disappear. You are done when every chart shows seven bars with error bars, Monday to Sunday.
* [x] **Check before you conclude** (≈ 20 minutes) with the checklist from the session. Ask the agent, then look yourself:
* Does the number of rows match the CSV, and are the values plausible?
* Are the weekdays in order, Monday to Sunday, and how many days stand behind each bar?
* What do the error bars show?
* Does the pattern hold in every product group?
* Does the chart answer the operations manager’s question?
How to read the intervals: if the intervals of two weekdays do not overlap, the difference is clear; if they overlap, the data cannot tell. The interval is about the *average* day, not about tomorrow: single days vary much more.
* [x] **Clean up the notebook** (≈ 30 minutes) so that it runs from top to bottom. Paste this prompt:
```
Clean up week-02-weekdays.ipynb: add a short Markdown cell at the top that says which question the notebook answers, a plain-language note before each step, and remove cells we no longer need. Keep everything in this one notebook.
```
You are done when *Restart* and then *Run All* (both in the notebook toolbar) run without errors and show the same charts. Do not use `/tidy-notebooks`: it splits notebooks into modules and comes in week 9.
* [x] **Draft the memo** (≈ 1 hour). First update the course skills: in Codex, Copilot, Cursor or another agent, paste the install prompt from [week 1](/opencampus-machine-learning-program/courses/einfuehrung/week-01-introduction.md) into your agent again; Claude Code updates them when you restart it. Then paste this prompt into your agent (in Codex, write `$audience` instead of “the audience skill”) and rewrite the draft yourself. The action title and the recommendation are yours.
```
Draft a decision memo for the operations manager from my weekday analysis in week-02-weekdays.ipynb. Follow the audience skill: an action title that states the conclusion, the answer first, at most three rounded numbers, how sure we can be, and one thing the data cannot tell. The data has no currency, so compare instead (for example "twice as much as on ..."). Include the one chart that supports the recommendation, saved as week-02-weekdays.png. Write the memo to week-02-memo.md: at most half a page, no jargon.
```
An action title states the conclusion, not the topic: “Rainy mornings need a third fewer bikes in the city centre”, not “Morning rentals by weather”. Do not use `/to-memo` (`$to-memo` in Codex) yet: it needs reviewed results, so it stops in week 2. It comes in week 6.
* [x] **Watch the other videos** listed below (≈ 47 minutes) and post your questions in the Element course room by the day before the next session.
**Stuck, or has your agent hit its usage limit?** Post your question in the Element course room. On a free plan, spread the tasks over two evenings.
{% hint style="info" %}
**Deliverable for the operations manager:** `week-02-memo.md` in your practice repository: at most half a page, with one chart, an action title, a recommendation for production planning per weekday and how sure we can be. Bring it to the next session; we discuss the memos in class.
{% endhint %}
### Videos and materials
**Slides:** [session 2 slides (PDF)](https://github.com/opencampus-sh/edu-ml-courses/tree/main/.gitbook/assets/intro-ds-week-02-slides.pdf)
**Example code:** [example plots](https://github.com/opencampus-sh/intro-to-data-science-and-ml/blob/main/02_visualization/example_plots.ipynb) with seaborn’s car data (`mpg`): a histogram, a bar chart with confidence intervals (the black lines), scatter plots and subplots you can reuse for the bakery data.
| Lesson / material | Source | Min |
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------------------------- | ------ |
| [Sorting](https://learn.deeplearning.ai/specializations/data-analytics/lesson/tm2ng/sorting) | Python for Data Analytics, M2 | 2 |
| [Sorting by multiple columns](https://learn.deeplearning.ai/specializations/data-analytics/lesson/wftuo/sorting-by-multiple-columns) | Python for Data Analytics, M2 | 2 |
| [Filtering](https://learn.deeplearning.ai/specializations/data-analytics/lesson/n5166/filtering) | Python for Data Analytics, M2 | 4 |
| [Filtering by multiple conditions](https://learn.deeplearning.ai/specializations/data-analytics/lesson/ngtd5/filtering-by-multiple-conditions) | Python for Data Analytics, M2 | 2 |
| [Selecting rows](https://learn.deeplearning.ai/specializations/data-analytics/lesson/f4q65/selecting-rows) | Python for Data Analytics, M2 | 5 |
| [Central tendency, variability, and skewness](https://learn.deeplearning.ai/specializations/data-analytics/lesson/a98w9/central-tendency%2C-variability%2C-and-skewness) | Python for Data Analytics, M2 | 4 |
| [Categorical data](https://learn.deeplearning.ai/specializations/data-analytics/lesson/uyl24/categorical-data) | Python for Data Analytics, M2 | 5 |
| [Correlation](https://learn.deeplearning.ai/specializations/data-analytics/lesson/zq7sk/correlation) | Python for Data Analytics, M2 | 5 |
| [Segmentation by one feature](https://learn.deeplearning.ai/specializations/data-analytics/lesson/o6dap/segmentation-by-one-feature) | Python for Data Analytics, M2 | 2 |
| [Segmentation by multiple features](https://learn.deeplearning.ai/specializations/data-analytics/lesson/aj6bg/segmentation-by-multiple-features) | Python for Data Analytics, M2 | 4 |
| [Plotting with matplotlib](https://learn.deeplearning.ai/specializations/data-analytics/lesson/q65f7/plotting-with-matplotlib) | Python for Data Analytics, M3 | 6 |
| [Colors, grids, & saving plots](https://learn.deeplearning.ai/specializations/data-analytics/lesson/a3y7h/colors%2C-grids%2C-%26-saving-plots) | Python for Data Analytics, M3 | 3 |
| [Grouped column charts](https://learn.deeplearning.ai/specializations/data-analytics/lesson/rtn1b/grouped-column-charts) | Python for Data Analytics, M3 | 5 |
| [Plotting with Seaborn](https://learn.deeplearning.ai/specializations/data-analytics/lesson/u924r/plotting-with-seaborn) | Python for Data Analytics, M3 | 4 |
| [Box plots](https://learn.deeplearning.ai/specializations/data-analytics/lesson/u4smy/box-plots) | Python for Data Analytics, M3 | 4 |
| [Histograms](https://learn.deeplearning.ai/specializations/data-analytics/lesson/ikzty/histograms) | Python for Data Analytics, M3 | 4 |
| [Matplotlib subplots](https://learn.deeplearning.ai/specializations/data-analytics/lesson/l39lj/matplotlib-subplots) | Python for Data Analytics, M3 | 4 |
| [Understanding confidence intervals](https://www.youtube.com/watch?v=tFWsuO9f74o) | YouTube (Dr Nic) | 4 |
| **Total** | | **69** |
### Optional
* The remaining videos of Python for Data Analytics, Module 3: text and annotations, ticks and spines, stacked column charts, scatter plots, method chaining, themes and palettes, other charts, combining charts, looping with subplots, pairplots (≈ 35 min).
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://opencampus.gitbook.io/opencampus-machine-learning-program/courses/einfuehrung/week-02-import-and-visualization.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.