Scribd
Upload
3 views2 pages

AIML Task 13

The document outlines an AI & ML internship task focused on PCA for dimensionality reduction using the MNIST or Sklearn digits dataset. It provides a step-by-step guide for implementing PCA, including data preparation, model training, and accuracy comparison, along with deliverables like variance plots and reports. Additionally, it includes submission guidelines and interview questions related to PCA concepts and practices.

Uploaded by

srikanthsri357951
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views2 pages

AIML Task 13

The document outlines an AI & ML internship task focused on PCA for dimensionality reduction using the MNIST or Sklearn digits dataset. It provides a step-by-step guide for implementing PCA, including data preparation, model training, and accuracy comparison, along with deliverables like variance plots and reports. Additionally, it includes submission guidelines and interview questions related to PCA concepts and practices.

Uploaded by

srikanthsri357951
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

AI & ML INTERNSHIP

✅ Task 13: PCA – Dimensionality Reduction


Dataset:
Primary: MNIST dataset
Alternative: Sklearn digits dataset

Tools:
Python
Scikit-learn
Matplotlib
Alternatives: OpenCV PCA
Hints / Mini Guide:

1. Load MNIST/digits dataset and flatten images into feature vectors.


2. Scale features using StandardScaler for correct PCA performance.
3. Apply PCA with multiple component counts like 2, 10, 30, 50.
4. Track explained variance ratio for each PCA setting.
5. Plot cumulative variance to choose best number of components.
6. Transform original dataset into reduced dimensional dataset.
7. Train Logistic Regression on reduced dataset.
8. Compare accuracy of original vs reduced dataset model.
9. Visualize PCA 2D scatter plot to observe separation.

Deliverables:

Explained variance plot


Reduced dataset
Accuracy comparison report

Final Outcome:
Intern understands feature compression and variance trade-off.

Interview Questions Related To Above Task:

What problem does PCA solve?


What is explained variance?
Why scaling required for PCA?
PCA vs feature selection?
What are PCA limitations?
📌 Task Submission Guidelines
⏰ Time Window:
Youcan complete the task anytime between 10:00 AM to 10:00 PM on the given day. Submission link closes at
10:00 PM
🔍 Self-Research Allowed:
Youare free to explore, Google, or refer to tutorials to understand concepts and complete the task effectively.
🛠 Debug Yourself:
Tryto resolve all errors by yourself. This helps you learn problem-solving and ensures you don’t face the same
issues in future tasks.
💸 No Paid Tools:
Ifthe task involves any paid software/tools, do not purchase anything. Just learn the process or find free
alternatives.
📁 GitHub Submission:
Create a new GitHub repository for each task.
Add everything you used for the task — code, datasets, screenshots (if any), and a short [Link] explaining
what you did.
📤 Submit Here:
After completing the task, paste your GitHub repo link and submit it using the link below:
👉
[Submission Link ]

You might also like