Iris- A small classic dataset from Fisher, 1936.

=========================

Classification problem

---

Out of three classes two are linearly separable and one is separable non linearly.

Dataset Information

Each instance is a plant

Additional information

This is one of the earliest datasets used in the literature on classification methods and widely used in statistics and machine learning. The data set contains 3 classes of 50 instances each, where each class refers to a type of iris plant. One class is linearly separable from the other 2; the latter are not linearly separable from each other.

Evaluation score :

Seems Overfitting , need to check what else can improve model accuracy score.

Test :---- When deleted features based on correlation heat map of independent feature.

recall_score: 71%

f1_score: 69%

precision: 71%

Test ----included all independent features.

recall_score: 89%

f1_score: 90%

precision: 90%

---

---

Train ----When deleted features based on correlation heat map of independent feature.

recall_score: 95%

f1_score: 95%

precision: 96%

---

Train ----included all independent features.

recall_score: 100%

f1_score: 100%

precision: 100%

---

---

Exploratory Data Analysis

Check Data Imbalance

Features histogram plot

Correlation checks

Check for Outliers

Pairplot to find independent features relation with each other

Overall status of the project: in progress

completed:

1. Data ingestion
2. Data validation
3. Data Transformation
4. Model training a. compare different model b. find best parameters and model
5. Model Evaluation
6. UIUX added

author:

Brajesh kumar