DocGarbanzo
DocGarbanzo
You need to install TF 2.2.0. There is a new section in the docs how to do this on windows: https://docs.donkeycar.com/guide/host_pc/setup_windows.html - can you check your error disappears with that?
Closed due to inactivity.
@cloud-rocket, @sctse999 - this makes perfect sense to me. It requires to simplify the part interface and interaction between parts - I made a mock design of a possible solution...
@cloud-rocket - I think this is too complicated for third party parts. We want users to write their own python classes, but we don't need them on pypi. They would...
> We are currently running a webserver on the car with an api that allows us to command the car through a rest API and/or websocket. We have an api...
Actually that is surprising because in DC 4.x we are using `tf.data.Dataset.from_generator()` which is supposedly the fastest way to shove data into the GPU and increase utilisation. I'm running a...
Looks like your miniconda base environment is somehow messed up. Can you re-install that? I would also recommend to run ```bash conda install mamba mamba env create -f install/envs/mac.yml ```...
Closed due to inactivity.
I get full CPU utilisation when running inference on RPi, because `numpy` and probably `tf` too are using multithreading internally.
Actually after a recent RPi software upgrade, tf has become much slower for '.h5' models, that's probably the reason why you are seeing this. Do you see the same affect...