Aleksei Rutkovskii

**I have exactly the same error.** ``` Traceback (most recent call last): File "categoricalWritersRunner.py", line 25, in df.parse_all_categories() File "/home/ubuntu/lightyear/flask_app/scripts/PeriodicWriters/weekly.py", line 76, in parse_all_categories self.parse_category(category) File "/home/ubuntu/lightyear/flask_app/scripts/PeriodicWriters/weekly.py", line 67, in...

@PeeInTheWind Hi, did you get it to work?

@jacobbieker Thank you for your help. I advanced forward a bit. 1. Your updates did make the code runnable. I obtained server with GPU that has 40 GB VRAM and...

Does anybody have answers to these questions? Thank you in advance! 1. How large are the GPU(s) on the machine on which the training of DGMR was done? 2. Has...

@jacobbieker Was 48GB of GPU memory enough to do training with 6 generation steps? And what was the issue with spectral norm? Was it making the model metrics become `nan`?...

Thank you for the information. I soon will try to train on multiple gpus Leaving solution and explanation for `nan` for people in case they encounter the same problem: **Explanation:**...

@hyungting Hi, I haven't analyzed the data yet, so I'm unsure of its format. Often, such datasets contain weather snapshots, which might be expressed in dBZ (Decibel relative to Z)....

@hyungting Okay. I understand! I am just looking into research that is done and the dataset used here [Skillful Precipitation Nowcasting Using Deep Generative Models of Radar](https://www.nature.com/articles/s41586-021-03854-z) and noticed that...

@hyungting Makes sense. Thank you so much for letting me know. Have a great day!

@hyungting Have you ran openclimatefix’s code on your machine? If so, could you please share the versions of the dependencies that you have used to run it? They do not...