| tags: | |
| - keras | |
| - time-series-classification | |
| - education | |
| # Wearable Activity Classifier – Group ___ | |
| ## Task | |
| Classify a 100-step, one-feature sensor sequence into **Stationary**, **Walking**, or **Running**. | |
| ## Model selected | |
| - Architecture: [CNN / SimpleRNN / LSTM / CNN+LSTM] | |
| - Input shape: `(100, 1)` | |
| - Output classes: 3 | |
| - Parameters: ______ | |
| ## Training data | |
| Synthetic signals generated in the class notebook. The dataset was designed for teaching and is not a real wearable benchmark. | |
| ## Evaluation | |
| - Test accuracy: ______ | |
| - Training time in our run: ______ seconds | |
| ## Why we selected this model | |
| [Write 2–4 sentences using evidence from your comparison.] | |
| ## Limitations | |
| - Synthetic, simplified data | |
| - One sensor feature only | |
| - No testing across real users/devices | |
| - Not intended for health, safety, or production use | |
| ## Team learning note | |
| [State one thing your group learned by comparing CNN, RNN, and LSTM.] | |