Vendor-neutral lab template

[profvjreddi]

For the labs, there are many different hardware solutions to pick from. we want to have one generic template that everyone follows so that vendors can produce their work according to that. Need to put together something like this that allows everyone to do their own thing but at the same time maintain consistency.

Lab Title

Provide a clear and concise title for the lab.

Objective

• Describe the main goal(s) of the lab.
• Explain what the student will learn or achieve by completing the lab.
• Specify the relevant section or chapter in the book this lab ties into.

Prerequisites

• List any necessary background knowledge or skills.
• Mention any required software, tools, or prior labs.
• Reference the specific reading parts of the book that should be reviewed before starting the lab.

Hardware and Software Requirements

• Detail the specific hardware needed for the lab.
• Include software versions and any additional dependencies.

Setup Instructions

• Step-by-step instructions for setting up the hardware and software.
• Include screenshots or diagrams if necessary.
• Troubleshooting tips for common setup issues.

ML System Lab Workflow

Data Collection and Preparation

• Data Source: Specify where the data is coming from (e.g., public dataset, simulated data, proprietary dataset).
• Data Description: Brief description of the data, including its size, format, and key attributes.
• Data Preprocessing Steps: List and explain the preprocessing steps required (e.g., cleaning, normalization, augmentation).

Model Training

• Model Architecture: Specify the architecture being used (e.g., CNN, RNN, Transformer).
• Framework: Mention the ML framework used (e.g., TensorFlow, PyTorch).
• Training Configuration: Detail the training configuration (e.g., batch size, learning rate, number of epochs).
• Code Snippets: Provide relevant code snippets for training the model.

Model Optimization

• Hyperparameters: List the hyperparameters tuned and their optimal values.
• Optimization Techniques: Describe any optimization techniques used (e.g., Grid Search, Random Search, Bayesian Optimization).
• Performance Metrics: Specify the metrics used to evaluate model performance (e.g., accuracy, F1 score).

Model Evaluation

• Evaluation Metrics: Detail the metrics used to evaluate the model (e.g., precision, recall, AUC-ROC).
• Validation Strategy: Describe the validation strategy (e.g., k-fold cross-validation, train-test split).
• Results: Present the evaluation results, including any visualizations if applicable.

Model Deployment

• Deployment Environment: Specify where the model will be deployed (e.g., cloud, edge device).
• Deployment Steps: Step-by-step instructions for deploying the model.
• Verification: Methods to verify the deployed model is working correctly.

Assessment

• Practical tasks to test students' understanding.
• Include answers or a separate section for solutions.

Summary

• Recap the main points covered in the lab.
• Highlight key takeaways.
• Mention any further reading in the book that could reinforce the lab's concepts.

References

• Cite any external resources or documentation referenced in the lab.
• Provide links to additional reading materials or related resources.

Appendix

• Any additional information, such as advanced configuration options or extended explanations.

[profvjreddi]

We might also want to have some sort of matrix at the start that shows people which of these devices do what like this: