How do I specify and search for optimal row and column sampling ratios in scikit learn?

Question

While learning random forrest I came across following points:

• Original dataset D contains d columns and n rows
• Each bootstrapped dataset Di contains