model_type string | layers list | activation string | optimizer string | learning_rate float64 | epochs int64 |
|---|---|---|---|---|---|
feedforward_regression | [
128,
64,
32
] | relu | adam | 0.001 | 50 |
Humanoid Energy Consumption Estimator (HECE)
Objective
Estimate energy consumption (kWh) during humanoid task execution.
Problem Type
Supervised Regression
Input Features
- payload_kg
- torque_index
- execution_time_s
- movement_speed_m_s
- ambient_temperature_c
Output
- predicted_energy_kwh
Model Architecture
- Input feature encoder
- 3-layer feedforward neural network
- Linear regression head
Training Configuration
- Loss: MSE
- Optimizer: Adam
- Learning rate: 0.001
- Epochs: 50
Evaluation Metrics
- MAE
- RMSE
Deployment Scenario
- Energy cost forecasting
- Efficiency optimization systems
- Operational budgeting tools
License
MIT
- Downloads last month
- 16