[英]Can I use reinforcement learning in tensorflowjs?
I'm doing a project at the moment which would require tensorflowjs to create a neural network that learns from reinforcement learning algorithms. 我目前正在做一个项目,需要tensorflowjs创建一个从强化学习算法中学习的神经网络。 Is this possible to do?
这可能吗?
Yes! 是!
There are many examples of this online. 网上有很多例子。 I would recommend you to take a look at https://github.com/BeTomorrow/ReImproveJS a reinforcement learning framework made using tfjs.
我建议您看看https://github.com/BeTomorrow/ReImproveJS一个使用tfjs制作的强化学习框架。
Example using ReImproveJS can be found here: https://gist.github.com/RGBKnights/756b5f51465cc22d0ca39205979ad2a1 可以在以下位置找到使用ReImproveJS的示例: https ://gist.github.com/RGBKnights/756b5f51465cc22d0ca39205979ad2a1
Also another example of applied tfjs is found here: https://www.metacar-project.com/ 此处还找到了另一个应用tfjs的示例: https ://www.metacar-project.com/
Another example of DQN applied to snake in tfjs, with instructions on how to run the code: https://github.com/tensorflow/tfjs-examples/tree/master/snake-dqn DQN在tfjs中应用于蛇的另一个示例,其中包含有关如何运行代码的说明: https : //github.com/tensorflow/tfjs-examples/tree/master/snake-dqn
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.