How one-eyed-warrior/qsvm works
Project type and purpose
• This is a small, notebook‐driven quantum computing mini project for coursework (21CSE306J).
• It loads tabular datasets (X_train_4000.txt, Y_train_4000.txt, X_test_4000.txt, Y_test_4000.txt, training_observables.csv), processes them, builds and trains a quantum (or hybrid quantum–classical) model in a Jupyter notebook, and then evaluates results.
File structure
• Top level:
• README.md – project description
• code.ipynb – the sole orchestration script, containing data ingestion, preprocessing, model definition, training, and evaluation cells
• *.txt and *.csv – raw input datasets
• No separate “src/” or “lib/” directories; the notebook itself implements every step.
README
• One‐line description: “Mini project for coursework in 21CSE306J Quantum Computing.”
a. Main components
• Data Layer: raw files (X_train, Y_train, X_test, Y_test, observables.csv)
• Preprocessing Module: notebook cells that load data (pandas/numpy), clean/transform as needed
• Quantum Model Module: cells that define quantum circuits (e.g., Qiskit or PennyLane), parameters, and trainer loop
• Execution Backend: quantum simulator or real quantum hardware API (e.g., IBM Q, local Aer simulator)
• Evaluation Module: notebook cells that compute metrics (accuracy, loss curves) and plot results (matplotlib/plotly)
• User Interface/Notebook: Jupyter environment hosting all of the above, interacting with the user
b. Relationships and interactions
c. Architectural patterns/design principles
• Pipeline pattern: clear sequential stages (ingest → process → compute → evaluate)
• Separation of concerns (data vs. model vs. evaluation), albeit all within one notebook
• External service integration (quantum hardware/simulator via API)
d. Key technologies/frameworks
• Python, Jupyter Notebook
• pandas, numpy for data handling
• Qiskit (or PennyLane) for quantum circuit definition and execution
• matplotlib or plotly for visualization
• IBM Q or other cloud‑based quantum backends (if used)
Elements
• Call out major libraries next to each component (e.g., “pandas/numpy” next to Preprocessing, “Qiskit” next to Quantum Model, “matplotlib” next to Evaluation)