Axel Sauer

This works for me (where cfg is a CfgNode): ``` from contextlib import redirect_stdout with open('test.yml', 'w') as f: with redirect_stdout(f): print(cfg.dump()) ```

answered in #31

aren't you the guy that made fun of our beautiful WordArt logo on Twitter? :joy: Finetuning is definitely possible, I'll fix this in the near future. Also, you can skip...

Actually, progressive growing can get you good results faster than standard training because the throughput is much higher at lower resolutions. Check out this figure from the OG progressive growing...

hey, sorry for the wait. I'm moved on to another project which needs my full attention atm. I still intend to adress the open issues, it might just take some...

Which version of timm are you using in your environment?

can you post the command you're using ?

Hey :) It is planned to release the code to produce the dataset + affordances by middle of May.

you can find examples of how to use THOR in the benchmarks ([vot.py](https://github.com/xl-sr/THOR/blob/master/benchmark/vot.py), [otb.py](https://github.com/xl-sr/THOR/blob/master/benchmark/otb.py)) or in [webcam_demo.py](https://github.com/xl-sr/THOR/blob/master/webcam_demo.py).

Hi all, the CARLA version is still 0.8.x. As @radeonwu already wrote, if you want to run the driving benchmark you need to go back to commit 36000a2 - this...