All the best in 2022 Phil, push out some more content for the 50k subs this year and I wish to have you on my show and not only talk about AI :) I think people are interested in you as a person as well! ;-)
Happy new year. I have one question please, Like I have one GNN RL model written in Pytorch, how can I plug or Integrate it with spark cluster, do you have any plan for making a video for this.
Hi Phil! I'm starting with RL to solve a custom environment created with gym. I've tried some RL frameworks, but I always struggle to make even little modification to the agent to test new things. Do you think I should stick more to RL frameworks and study them more deeply or try to be "more free" and create everything from zero ( example, using PyTorch only to handle the creation of the NN and the backpropagation ) ?
Hey Phil, been following for long time, I see you always use Linux (Ubuntu). I am thinking about migrating from my MacBook Pro 2017 to a some good hardware and run Ubuntu on top of it. However, I have seen many battery concerns regarding linux distributions, what are your thoughts?
Honestly, I always use desktops. I can't comment on battery life for laptops. I'd recommend making the jump regardless. Since you're probably comfortable using the terminal, you should be fine in Linux.
Happy new year Vitalii. I didn't aim for any real learning for this demo. I get the feeling some real tweaking is needed. If you want, I can do a follow up video on hyperparameter tuning for ACME.
Loved it when you said "If you are running Windows, you should probably upgrade to a real operating system"
:D
Nice to know. This is something really exciting and I will like to see more videos on this. Thank you Phil. Happy new year to you.
Happy new year!
Happy New Year, Phil! Huge fan of your tutorials and courses. Keep ‘em coming.
Thank you Rofi!
I love you, as a teacher
Thank you Gav
Thank you phil! Happy new year :) keep the great videos coming
Happy new year!
Love sir. I learned so much from you thank you .
Keep uploading new videos to share knowledge with us .
Thank you Vishnu.
All the best in 2022 Phil, push out some more content for the 50k subs this year and I wish to have you on my show and not only talk about AI :) I think people are interested in you as a person as well! ;-)
Thank you Jousef. I will push out more content and ping you about appearing.
Many thanks for the video !!!!
Happy new year. I have one question please, Like I have one GNN RL model written in Pytorch, how can I plug or Integrate it with spark cluster, do you have any plan for making a video for this.
Honestly, I have no experience with spark.
Hi Phil! I'm starting with RL to solve a custom environment created with gym. I've tried some RL frameworks, but I always struggle to make even little modification to the agent to test new things.
Do you think I should stick more to RL frameworks and study them more deeply or try to be "more free" and create everything from zero ( example, using PyTorch only to handle the creation of the NN and the backpropagation ) ?
Hey Phil, been following for long time, I see you always use Linux (Ubuntu). I am thinking about migrating from my MacBook Pro 2017 to a some good hardware and run Ubuntu on top of it. However, I have seen many battery concerns regarding linux distributions, what are your thoughts?
Honestly, I always use desktops. I can't comment on battery life for laptops. I'd recommend making the jump regardless. Since you're probably comfortable using the terminal, you should be fine in Linux.
Happy NY!
Don't you achieve any results with training Space Invaders? How many episodes did you train SI agent?
Happy new year Vitalii. I didn't aim for any real learning for this demo. I get the feeling some real tweaking is needed. If you want, I can do a follow up video on hyperparameter tuning for ACME.
@@MachineLearningwithPhil It would be great.
Sir I need new video from you:
Sir try to code ppo with tensorflow 2.o or keras plzzz🥺🥺🥺🥺🥺🥺
dm-reverb is not getting installed in Win11 python 3.7. What to do?
Reverb is only supported on Linux based operating systems. Maybe try the Linux subsystem for Windows.
what do you think of ray rllib ?
I have only taken a brief look at it, but it looks like a great framework. It's probably very useful for getting advanced algorithms working quickly.
Can you share all papers
Dueling DQN: arxiv.org/abs/1511.06581
DDPG: arxiv.org/abs/1509.02971
@@MachineLearningwithPhil thanks sir
Is this still not working with MacOS?
Haven't tried it. It probably does, considering how close it is to Linux
Phil > You should upgrade to a real operating system
Linux users > Exactly
Windows users > 🥲