Riya Shukla

Tech student | Intern @InamigosF

[Riya Shukla @ria0106_here]

Day 26 of learning ML 📊

Today I learned:
• Percentile Method for outlier handling
• Winsorization
• Feature Construction
• Feature Splitting
• How engineered features impact ML models

#MachineLearning #DataScience #100DaysOfML

[Riya Shukla @ria0106_here]

Wrapped up the session with hands-on practice to understand when and how to handle outliers effectively.

#MachineLearning #DataScience #100DaysOfML #WomenInTech

[Riya Shukla @ria0106_here]

Day 25 of learning ML 📊

Today I learned:
• What outliers are
• Their effect on ML models
• Detection methods
• Trimming & Capping
• Z-Score Method

[Riya Shukla @ria0106_here]

Day 24 of learning ML

Today I explored multivariate imputation techniques:
• KNN Imputation
• MICE (Multiple Imputation by Chained Equations)
Learned how these methods use relationships between features to estimate missing values more effectively than simple imputation methods

[Riya Shukla @ria0106_here]

Day 23 of learning ML 📊

Today I learned:
• Random Value Imputation
• Missing Indicator Technique

Explored how these methods handle missing values while retaining useful information for machine learning models.

#MachineLearning #DataScience #100DaysOfML

[Riya Shukla @ria0106_here]

Practiced different imputation techniques and learned how the choice of method depends on the data and the problem you're trying to solve.

Data preprocessing continues to prove that good models start with good data. 📊

#ML #DataScience #LearningInPublic #100DaysOfML

[Riya Shukla @ria0106_here]

Day 22 of my ML journey 🚀

Today's focus was handling missing data in machine learning.

Covered:
✅ Mean & Median Imputation
✅ Arbitrary Value Imputation
✅ End of Distribution Imputation
✅ Most Frequent Imputation (Categorical Data)
✅ Missing Category Imputation

[Riya Shukla @ria0106_here]

✅ Hands-on practice with real datasets

The more I learn, the more I realize that data preprocessing is where most of the real work in machine learning happens.

#MachineLearning #DataScience #FeatureEngineering #LearningInPublic #100DaysOfML

[Riya Shukla @ria0106_here]

Day 21 of my ML journey 🚀

Today's focus was handling real-world data before feeding it into ML models.

Covered:
✅ Date & Time feature handling
✅ Missing data handling
✅ Complete Case Analysis (CCA)
• Assumptions
• Advantages & disadvantages
• When to use it

[Riya Shukla @ria0106_here]

Learning how transforming raw data into meaningful features can improve model performance. 🚀

#MachineLearning #DataScience #100DaysOfML

[Riya Shukla @ria0106_here]

Day 20 of learning ML 📊

Today I explored:
• Binning & Binarization
• Numerical feature encoding
• Uniform, Quantile & K-Means binning
• Encoding discretized features
• Custom/Domain-based binning
• Mixed data handling

[Riya Shukla @ria0106_here]

Seeing how pipelines make preprocessing more organized, reusable, and less error-prone gave me a better appreciation for writing scalable ML code.

#MachineLearning #DataScience #ScikitLearn #LearningInPublic #100DaysOfML

[Riya Shukla @ria0106_here]

Day 19 of #100DaysOfML🚀

Today I learned how to build cleaner and more efficient ML workflows using Column Transformers and Pipelines:

✅Column Transformer
✅Challenges of non-pipelined preprocessing
✅Building preprocessing pipelines
✅Hands-on pipeline implementation

[Riya Shukla @ria0106_here]

@takeshi7_ Hey there....Let's connect

[Riya Shukla @ria0106_here]

@Ansh14225311 Let's be mutuals

[Riya Shukla @ria0106_here]

Day 18 of learning ML 📚

Today I learned how to manage categorical data:
• Ordinal Encoding
• Nominal Encoding
• One-Hot Encoding

Applied these techniques on a cars dataset to understand their practical use.

Slowly building a stronger foundation in data preprocessing

[Riya Shukla @ria0106_here]

Together, we can create measurable change, empower lives, and build a more inclusive future.

#CSR #SocialImpact #ESG #CommunityDevelopment #Sustainability #InAmigosFoundation

[Riya Shukla @ria0106_here]

🤝 Turn your CSR vision into meaningful impact.

Partner with InAmigos Foundation to support initiatives in 📚 education, 👩‍🎓 women empowerment, 🌱 environmental sustainability, 🐾 animal welfare, and 🤝 community development.

[Riya Shukla @ria0106_here]

🔹 MaxAbs scaling → for sparse data
🔹 Robust scaling → when you have outliers

Practised on the Wine dataset. Same data. Different scales. Different results. 🧠

Scaling isn't just preprocessing. It's essential. 🚀

#100DaysOfML #MachineLearning #DataScience #Python

[Riya Shukla @ria0106_here]

Day 17 of #100DaysOfML 🧵

Your features are probably on different scales.

One column might be 0-100. Another 0-10000.

Your model will get confused. 😅

Today I learned feature scaling:
🔹 MinMax scaling → 0-1 range
🔹 Mean normalization → centering data