[UPDATE] Unit 1, sampling from observation space

[dfomin]

The observation space is 8 float numbers but we know from the documentation that the last 2 numbers are actually boolean so they can be only 0 or 1. In a line
print("Sample observation", env.observation_space.sample()) # Get a random observation
we have 8 random floats including the last 2 of them and I found it a bit misleading since we can never get this observation from the environment. Maybe some additional clarification is needed here.