Hi bro,liked your video video very much.I tried the Cart Pole of openAI gym in my system and trained it for a day.But it still falls after some 425 steps ,wont it learn to balance completely?
Great video, I had it playing breakout & Pong. I tried to implement Pickle to save the games as a binary, so I can stop & re-start the game from the same learning point. But I was not successful. Would you know how to save the Q table so it doesn't have to start from scratch every time?
Keep it up bro can't wait for future RL videos
Thanks Jonas, That's really great to hear!
Hi bro,liked your video video very much.I tried the Cart Pole of openAI gym in my system and trained it for a day.But it still falls after some 425 steps ,wont it learn to balance completely?
love to see more openai gym projects: )
Great tutorial!
Great video sir but you are mistaken in thinking that background music is wanted or needed.
Impressive! Keep it up
I love your videos, thank you for your content
Thank you! And my pleasure!
This is great. But it would be even better with no music. It is really distracting when trying to learn complex environments like double dqns.
I actually like it. It is stress free hehe.
Great video!
great!!
Hey I want to make a bot for a 3D game what data should I feed it
You make it more complex for beginners by combining multiple games. Should have focus on one game
My Q Learning with neural network was worse than simply using hillclimb...
Like the content, hate the music
Why when CNN added to this everything crashes, the agent stops learning??
Huhh... You are so fast ...my brain got stuck to digest all together .. take deep breath.. May request to go slow .
Great video, I had it playing breakout & Pong. I tried to implement Pickle to save the games as a binary, so I can stop & re-start the game from the same learning point. But I was not successful. Would you know how to save the Q table so it doesn't have to start from scratch every time?
Output it into a CSV file?