Learn2Hack Tutorials
Learn2Hack Tutorials helps you master data analysis and build AI/ML solutions in Python through clear, practical, and hands-on lessons.
Prerequisite Knowledge
Recommended (Not All Strictly Required):
-
Basic Python: variables, loops, functions, imports.
-
High‑School Algebra: linear equations, basic functions.
-
Basic Statistics: mean, variance, distributions, train/test split.
-
Data Literacy: reading CSVs, handling missing values.
-
High-School Biology: helpful for interpretation.
PyTorch (Basics)
You only need a minimal subset of PyTorch to participate: tensors, basic operations, building a small nn.Module, forward pass, loss, and an optimizer loop. You can learn deeper topics (custom datasets, schedulers, mixed precision) later if your project demands it.
Focus on: torch.tensor, shapes, nn.Linear, nn.Sequential, optimizer.step(), and monitoring loss.
Pandas (Data Handling)
Pandas lets you explore, clean, join, and slice tabular biomedical data quickly.
-
Load Data: pd.read_csv()
-
Inspect: df.head(), df.info(), df.describe()
-
Filter/Select: boolean masks, loc, iloc
-
Group & Aggregate: groupby, agg
-
Clean: Handling Missing Values (dropna, fillna)
Mastering these covers 80% of what you will do in early experiments.
Notebooks & Environment
We recommend VS Code with the official Jupyter extension for iterative exploration plus a standard project folder for scripts & modules you refactor later. Alternatively, try out Google Colab for a cloud-based Jupyter experience with GPU access.
1. Install VS Code
2. Add the Python + Jupyter extensions
3. Create a virtual environment (python -m venv .venv)
4. Open / Create a .ipynb Notebook and Select the venv kernel
5. Incrementally Prototype → Migrate Stable Code into .py files.
Notebooks are for exploration; keep final training/evaluation pipelines scripted for reproducibility.
