----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv1d-1 [-1, 6, 25] 96
ReLU-2 [-1, 6, 25] 0
Dropout-3 [-1, 150] 0
Linear-4 [-1, 30] 4,530
ReLU-5 [-1, 30] 0
Linear-6 [-1, 2] 62
================================================================
Total params: 4,688
Trainable params: 4,688
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.00
Forward/backward pass size (MB): 0.00
Params size (MB): 0.02
Estimated Total Size (MB): 0.02
----------------------------------------------------------------
- Add BLE connectivity to esp32
- Use nimble instead
- Train model
- Deploy model onto esp32
- Apply DSP algs to make the prediction more stable