Hi im a senior working with a team on a digital of a UAV. I will preface by saying that i just started learning about deep learning. My task is to build a neural network (2 layer feed forward algo) model that will take sensor data as input and output the best simulation model for the physical asset. the sim models will be created with ANSYS and 3D modeled in unity. I have these 2 questions: 1. How should i format the output? I understand it should be one hot encoded to allow for multiple outputs representing each ANSYS model (basically, the best model representing the status of the asset will have the highest activated node, we do the association manually, e.g. output 1 = model 1, output 2 = model 2, etc) How would I go about doing that? 2. What should the input be? If i gather temperature, speed, vibration, stress, micro-tears, etc as input, how do i feed it to the model? since every feature will be tables with thousands of rows (ticks in time, value) per model, should i take the average of every feature and use those as inputs? Thank you TC: 70k as a Summer intern Yoe: <1 #tech #digitalasset #machinelearning #machinelearningengineer #datascience
Tech Industry
Yesterday
2950
What happens when most of your team is Indian?
Tech Industry
Yesterday
260
Which degree is more worth it?
Health & Wellness
Yesterday
610
Lasik cost
India
2h
225
'Hindutva': The Radical Hindu Ideology That Seeks to 'Push Christianity Out of India’
Tech Industry
Yesterday
1667
Women, help me understand why this is inspirational