This project is a machine learning workshop organized by SHE CODE. It aims to introduce participants to the fundamentals of machine learning and provide hands-on experience with various ML algorithms and techniques. I learned two new features: YData Profiling, used in data analysis, and AutoGluon, used for model prediction.
To get started with this project, follow these steps:
- Clone the repository:
git clone https://github.com/Bbrnn/SHE-CODE-ML-WORKSHOP.git
One of the key components of this workshop is the use of YData Profiling for exploratory data analysis. YData Profiling is a powerful tool that provides comprehensive insights into your dataset, helping you understand its structure, identify missing values, outliers, and much more. To use YData Profiling, follow these steps:
- Install YData Profiling:
pip install ydata_profiling
- Import the library in your Python script:
from ydata_profiling import ProfileReport
- Load your dataset using pandas:
import pandas as pd; data = pd.read_csv('path/to/your/dataset.csv')
- Generate the profile report:
profile = ProfileReport(data); profile.to_file("your_report.html")
Another important aspect of this workshop is the use of AutoGluon for automated machine learning (AutoML). AutoGluon is a powerful AutoML toolkit that simplifies the process of training and tuning machine learning models. To use AutoGluon, follow these steps:
- Install AutoGluon:
pip install autogluon
- Import the library in your Python script:
from autogluon.tabular import TabularPredictor
- Load your dataset:
train_data = TabularPredictor.load_csv('path/to/your/train_data.csv')
- Define the target variable:
label = 'target_column_name'
- Train the model:
predictor = TabularPredictor(label=label).fit(train_data)
- Make predictions:
test_data = TabularPredictor.load_csv('path/to/your/test_data.csv')
predictions = predictor.predict(test_data)
Remember to replace 'path/to/your/dataset.csv'
, 'path/to/your/train_data.csv'
, and 'path/to/your/test_data.csv'
with the actual paths to your dataset files.