gym VS tensorflow

Good evening, guys. I currently use StarCraft 2 as a tool for experimenting with my deep reinforcement learning projects, I have also used OpenAI Gym.

> and focus instead on the 5-10% of the application that’s hyper specific to your industry / vehicle / customer. Just like our friends in SaaS.

Ah so the thing that actually takes 90-95% of the time? "Twillo" is no good to me if I dont integrate it properly.

I struggle to see what you've built here other than gym[0] with a largely useless API (Come on, ML training doesn't suck because of lack of HTTP API's), sure its hyperfocused on automating vehicle but thats something that's existed for a while - atleast for drones [1].

Shoving sensors on random devices doesn't work that easy, you know that - I dont need a PHD to tell you that. Farmers likely aren't building prod grade ML data sets (except those SWE who gave up FAANG after making $BANK)

What's the real value prop? Why shouldn't I remake the environment in Hym and just cut you out? Gym doesn't require reems of code for environments, same as you.

[0] https://www.gymlibrary.ml

This course (the first in a planned multi-part series) shows how to use the Deep Reinforcement Learning framework RLlib to solve OpenAI Gym environments. I provide a big-picture overview of RL and show how to use the tools to get the job done. This approach is similar to learning Deep Learning by building and training various deep networks using a high-level framework e.g. Keras.

We are making mujoco installation a lot easier (i.e. pip install gym[mujoco]) without all the pains with mujoco-py by adopting Deepmind’s new mujoco bindings https://github.com/openai/gym/pull/2762, but this is a work in progress…

And according to the source code, the frame stack returns the most recent observations.

I don't know if it is seamlessly compatible with this specific environment, but in general you can use the PixelObservationWrapper for this type of thing (with pixels_only=False). https://github.com/openai/gym/blob/master/gym/wrappers/pixel_observation.py

I would recommend doing this using an environment RewardWrapper. Here you have an example https://github.com/openai/gym/blob/master/gym/wrappers/transform_reward.py

See https://github.com/openai/gym/issues/2510

It all started with this issue.

What do you mean? TF is under active development https://github.com/tensorflow/tensorflow/commits/master

The meme doesn't specify if it's the interpreter or some popular library. It could be anything.

For example, there is PyTorch which is primarily C++ but has Python bindings. Most people use the Python bindings, same for TensorFlow. JAX is mostly Python, same for scikit-learn.

I am trying to run the freeze_graph.py git source directly from tensorflow, however no matter the configuration I cannot run the script and I recieve the following error...

None of the unity git projects I have found walk through the start to finish steps and most just use ONNX models taken from the zoo or a public source. I have seen some mention that I may need to 'freeze' my Saved_model.pb file, but nothing I have found works or explains how I might do this. Ether the code fails to run (such as Freeze Graph Python Script - which fails to import 'from tensorflow.python.checkpoint' no matter my configuration, OR requests args that I don't seem to have.)

Open-source projects leveraged by nebullvm include OpenVINO, TensorRT, Intel Neural Compressor, SparseML and DeepSparse, Apache TVM, ONNX Runtime, TFlite and XLA. A huge thank you to the open-source community for developing and maintaining these amazing projects.

...however, this project fails to read in the Unity Sample projects I've tried to use. From various posts it seems that I may need to use a 'frozen' save_model.pd before converting it to ONNX. However all the guides and python functions that seem to be used for freezing save_models require a lot more arguments than I have awareness of or data for after going through so many systems. Freeze_Graph python