Training a AI engine that takes video input is very resource hungry… therefore we’d like to process the bare minimum data, in the process of reducing the inputs an “autocrop” function becomes handy ! download : https://gist.github.com/michel47/8f60c14b55553a98d32d
While doing my TensorFlow tutorials this evening I ran across a possible solution to add randomness to the decision-making of our Project Pendulum AI bot. TensorFlow includes a stochastic optimization training technique called the Adam Optimizer algorithm. See the published paper on this method here.
Here is what we are trying to achieve : a human holds a pendulum pendulum amplifies hand’s micro-movements a machine watches a pendulum swinging data is analyzed to recover the human thought signature from the brain induced micro-movements of the hand However, instead of hard coding the algorithm, we…