jsphon / reinforcement_learning

Python Package For Reinforcement Learning
0 stars 0 forks source link

Analyse Performance of the line world example #44

Open jsphon opened 6 years ago

jsphon commented 6 years ago

The example is very simple, but very slow.

Find out why it is so slow.

Is it slow because of calculate states, or because of the neural network training? If it is due to the network training then we probably need a gpu.

If it is slow due to target generation, perhaps we can vectorize more calculations.