the aim of the project is to classify iris flowers amonge three species (setosa, versicolor, and virginica) from the measurements of sepal length and width and petal length and width. the iris datasets contain 150 samples of three classes of 50 each, where each class refers to a kind of iris specie.
the overall goal is to design a model that makes useful classifications after cleaning and trainig the model. we then use the attributes of the flower to predict the species of the iris flower.
the data source is the iris_flowers.datasets. it contains the data for this example.
git clone https//github.com/EHICHIOYA127/iris-classifier.git cd iris-classifier python -m venv venv && source venv/bin/activate pip install -r requirements.txt python src/train.py
-
Data gathering: collecting the iris data, which includes measurements of sepal length and width and petal length and width.
-
Data processing: Cleaning and preparing the data for training, including encoding variables, and splitting the data into training and testing sets.
-
Model Training: Choosing machine learning algorithm and training the model using the preprocessed data.
-
Making predictions: Here, the attributes of the flowers will be used to predict the species of the flower (i.e using the sepal length and width, and the petal length and width to predicts if an iris flower is of type Setosa, Versicolor, or virginica).
-
Evaluating the model: Assessing the model's performance using various metrics such as accuracy score and confusion metrix to guage its effectiveness in classifying iris species
-
Interpreting the model: Here, from the predictions made from the first 10 predicted speciess and the first 10 actual species, it could be observed that the first ten predicted values [1, 0, 2, 1, 1, 0, 1, 2, 1, 1], matches the first ten true values [1, 0, 2, 1, 1, 0, 1, 2, 1, 1]. this shows a good sign that the model is performing well. and also,with the accuracy score of 100%, it revealed that all predictions made were accurate. hence, the model achieved a perfect prediction accuracy in this project.
Feel free to submit pull requests or open issues.
MIT License
Email: ehichioyasunny@gmail.com