https://github.com/gperdrizet/ds-12
Course materials for 4Geeks Academy data science cohort 12
https://github.com/gperdrizet/ds-12
data-science python
Last synced: about 2 months ago
JSON representation
Course materials for 4Geeks Academy data science cohort 12
- Host: GitHub
- URL: https://github.com/gperdrizet/ds-12
- Owner: gperdrizet
- License: gpl-3.0
- Created: 2025-06-23T23:17:01.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-09-16T00:57:21.000Z (10 months ago)
- Last Synced: 2025-09-16T02:41:34.624Z (10 months ago)
- Topics: data-science, python
- Language: Jupyter Notebook
- Homepage:
- Size: 10.5 MB
- Stars: 3
- Watchers: 0
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ds-12
Course materials for ds-12
1. [YouTube playlist](https://youtu.be/607QEWYZQpU?si=rBIrfjwxsHJk3xf4)
2. [Module slides](https://github.com/gperdrizet/ds-12/blob/main/pages/slides.md)
3. [Project solutions](https://github.com/gperdrizet/ds-12/blob/main/pages/solutions.md)
4. [How-to guides](https://github.com/gperdrizet/ds-12/blob/main/pages/guides.md)
## Extras
### 2025-07-18
OpenAI just released their ChatGPT based agent yesterday - here are the details:
- Press release/FAQ style overview: [ChatGPT agent](https://help.openai.com/en/articles/11752874-chatgpt-agent)
- Full technical details: [ChatGPT Agent System Card](https://cdn.openai.com/pdf/839e66fc-602c-48bf-81d3-b21eacc3459d/chatgpt_agent_system_card.pdf)
### 2025-07-16
While we are on the 'math' portion of the course one good, if a little obscure, Python library to know about is [SymPy](https://www.sympy.org/en/index.html). It does symbolic math in Python - including derivatives. We won't run into it often, but its good to know its out there in case you ever need it. Here's and example from the documentation - calculating the first derivative of a cosine function:
```python
import sympy as sp
x = sp.symbols('x')
derivative = sp.diff(sp.cos(x), x)
print(f'First derivative: str(derivative)')
```
```text
First derivative: -sin(x)
```
### 2025-07-14
As promised here is an 'extra' assignment which will walk you through hard-coding your own optimizer in Python to fit a linear model to toy data. Highly recommend taking a look - the assignment will give you a good 'gut' feeling for what is happening under the hood when we train machine learning models:
[Linear Regression & Optimization Assignment](https://github.com/4GeeksAcademy/gperdrizet-optimization-bonus-assignment)
2024 Nobel prize in physics was awarded for early research which lead to modern neural networks. The prize was shared between two researchers: John Hopfield, who invented the 'Hopfield network' and Geoffrey Hinton, who designed early gradient descent algorithms.
1. [2024 Nobel Prize in Physics](https://www.nobelprize.org/prizes/physics/2024/popular-information/): description of the history and importance of the works
2. [ADAM: A METHOD FOR STOCHASTIC OPTIMIZATION](https://arxiv.org/pdf/1412.6980): Scientific paper describing ADAM, one of the most common/popular optimization algorithms for training neural networks (note the publication year and the first authors affiliations!).
### 2025-07-11
Interesting further topic to read up on while we are learning about APIs: [Model Context Protocol](https://modelcontextprotocol.io/introduction). MCP was originally proposed by Anthropic, but is an open standard that anyone can use. It's basically a type of API designed for LLMs and agents to use. It standardizes communication between the model and data source, allowing a way to easily use and share tools for building agents. See also [A2A](https://developers.googleblog.com/en/a2a-a-new-era-of-agent-interoperability/) (Google) and [ACP](https://www.ibm.com/think/topics/agent-communication-protocol) (IBM) - same idea, but for communication between agents.
### 2025-07-02
Cool talk by Bohan Zhang of OpenAI's infrastructure team - covers their implementation of PostgreSQL and shows what is possible with a cutting edge, production grade SQL database at a top company: [OpenAI: Scaling PostgreSQL to the Next Level](https://www.pixelstech.net/article/1747708863-openai%3a-scaling-postgresql-to-the-next-level).
### 2025-06-27
Useful Pandas methods for the real estate data cleanup assignment:
1. `.sort_values()` used to sort a dataframe
2. `.unique()` & `.nunique()` used to get information about unique values in a dataframe/series
3. `.isna()` checks for NaN (not a number) missing value placeholders
3. `.dropna()` used to remove NaN (not a number) missing value placeholder from a dataframe or series
You can find more information about what these methods do and how to use them in the Pandas [DataFrame](https://pandas.pydata.org/docs/reference/frame.html) and [general function](https://pandas.pydata.org/docs/reference/general_functions.html) documentation.
There is a whole module about plotting coming up - but for now, a quick skim of the Matplotlib [hist](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.hist.html) documentation should be enough to complete the last question.