Exploring Multi-Classification Confusion Matrix

•

Part 3️⃣ Exploring Multi Classification Confusion Matrix In Machine Learning 🟢 Generate confusion matrix using sklearn 🔵 Visualize confusion matrix using Seaborn & Matplotlib 🟠 Generate classification report using sklearn 🟡 Derive conclusion on which model does better

What Is Multi Classification ? 🔵 Classify instances into one of three or more distinct classes or categories. 🟡 Applicable to diverse scenarios such as image recognition, language processing, and medical diagnosis.

How Is It Different From Binary Classification ? 🔵 In Layman's terms we have more than 2 buckets to categorize items into 🟡 As a result simple True Positive, False Positive, True Negative & False Negatives described previously do not apply here twitter.com/_jaydeepkarale/status/1686000750902829056

🔵 Our dataset to explain Multi Classification are dummy model outputs from 2 models which try to classify animals. 🟡 It contains in one column the actual animal and in another the predicted value by the model

We then use the Seaborn & Matplotlib to plot the confusion matrix as a heatmap. Some observations we can see 🟢 Model 1 does a better job classifying most animals except 🐼 🔴 Model 2 fares better when classifying 🐼

Let's now view the `classification_report` for both models Some observations we can see 🟢 Model 1 has an overall better accuracy 🟢 Model 2 has better accuracy classifying 🐼 ✅ Using an ensemble learning approach combining `Model 1` & `Model 2` (for 🐼 ) may be better

References & dataset courtesy youtube.com/watch?v=zqJLLaDd4QQ

All Code available on GitHub It costs you 0 to leave a ⭐ :) github.com/jaydeepkarale/100DaysOfMachineLearning/blob/main/multiclassification_confusion_matrix.ipynb

Hello 👋 I am Jaydeep from India 🇮🇳 Full time Software Engineer & part time content creator on 🐦Twitter 🖧 Linkedin 🎥YouTube Follow me for content on 🐍 Python 🤖Ai/ML 🎨Data Visualization 🌟Content creation Subscribe To My YouTube🔽 youtu.be/FLdS-kBt88M