I want to clarify following questions on Neural Network:
- Why do we have 3 units at the minimum in first layer in the tensorflow?
- How do we know how many layers in total we need in tensorflow?
I want to clarify following questions on Neural Network:
If this question is related to a specific course, please post the question in the “Course Q&A” area for that course.
General replies:
The number of units in the input layer must match the number of features in each example in the data set.
The number of layers needed is determined by experimentation. More layers allows for a more complex model, but it also requires a lot more resources for training.
Thanks @TMosh. Appreciate your help with this!