Assignment: End$-$to$-$End Machine Learning Workflow with KizenML, XAI, and Cloud

Deployment

Total Marks: $30$

Objective:

In this assignment, you will build a complete end$-$to$-$end machine learning $($ML$)$

workflow using KizenML and Explainable AI $($XAI$)$ tools. You are free to choose any

other tools or frameworks that support the workflow, including cloud services like

AWS or Azure for deployment. The final deliverables include a screen recording

explaining your process and detailed documentation.

Tasks and Marks Distribution:

$1.$Data Collection and Preprocessing $(4$ Marks$)$:

$$Task: Select an appropriate dataset and perform data preprocessing, including data

cleaning, feature engineering, and scaling$/$normalization$.$

$$Details: Explain the choices made during preprocessing and how they impact the

model. Use KizenML or other tools for AutoEDA if applicable.

$$Marks: $4$ Marks

$2.$Model Selection, Training, and Hyperparameter Tuning $(6$ Marks$)$:

$$Task: Train multiple models, tune hyperparameters, and select the best$-$performing

model.

$$Details: Utilize tools like AutoML, KizenML, or others for model selection and

hyperparameter tuning. Document the experimentation process and justify your

model choice.

$$Marks: $6$ Marks

$3.$Explainable AI $($XAI$)$ Implementation $(6$ Marks$)$:

$$Task: Apply Explainable AI techniques to make your model$$s predictions

interpretable.

$$Details: Use XAI tools such as SHAP, LIME, or others to provide insights into your

model$$s decision$-$making process. Discuss the importance of interpretability in your

model and how XAI tools helped achieve it$.$

$$Marks: $6$ Marks

$4.$Model Deployment Using Cloud Services $(6$ Marks$)$:

$$Task: Deploy your trained model using free$-$tier cloud services such as AWS or

Azure.

$$Details: Create an API or web interface using services like AWS Lambda, Azure

Functions, or any other cloud service. Demonstrate how to make predictions using

the deployed model.

$$Marks: $6$ Marks

$5.$Workflow Documentation and Explanation $(8$ Marks$)$:

$$Task: Document the entire end$-$to$-$end workflow and provide a screen recording that

explains the process.

$$Details:

$$Screen Recording $(4$ Marks$)$: Record a video walkthrough of your workflow,

explaining each step and the tools used.

$$Documentation $(4$ Marks$)$: Provide detailed written documentation that includes

code snippets, explanations of each step, challenges encountered, and how they

were resolved.

$$Marks: $8$ Marks

Deliverables:

$$Screen Recording: A video explaining the end$-$to$-$end workflow, from data collection

to model deployment and XAI implementation.

$$Code Repository: Submit a well$-$organized repository containing all code, including

scripts for data preprocessing, model training, XAI implementation, and deployment.

$$Documentation: A detailed report that covers each step of the workflow, with

explanations, challenges, and solutions.