Pyroluk

Thx to wiringpi, this module cannot be used on 64bit versions of Raspbian. This module, on the other hand, works flawlessly on 64bit OS versions: https://github.com/ryck/MMM-DHT-Sensor

Thank you for taking time out of your day to have a look on my problem. I can happily report, that I solved it by fixing my externalized standard scaling...

Hi, I have a similar problem. Using the latest TFT version in TSPP (Docker) and enabling AMP, I get the following error: ``` ValueError: Error instantiating 'training.trainer.CTLTrainer' : Attempted to...

Hi, got it kind of working with the dummy_batch code and the TFT_SCRIPTING=False flag. For other readers, the self.train_dataloader init code needs to be moved above that code in order...

Had to reduce the batch size for a quick run on my local 8GB GPU: `TFT_SCRIPTING=false` `model=tft dataset=electricity trainer/criterion=quantile trainer.config.num_epochs=1 trainer.config.amp=true trainer.config.batch_size=768` Results: `MAE : 0.2200114130973816 RMSE : 0.3554209768772125 SMAPE...

Fresh repro pull and docker build solved the problem. `TFT_SCRIPTING=false python launch_training.py model=tft dataset=electricity trainer/criterion=quantile trainer.config.num_epochs=1 trainer.config.amp=true trainer.config.batch_size=768` Results: `MAE : 45.368255561589564 RMSE : 368.65816141232375 SMAPE : 9.376580496996748 ND :...

No joke, it just used up all of my 64GB(!!!!!) of RAM on my windows machine. And I am running the normal Nvidia PyTorch containers (nvcr.io/nvidia/pytorch). SBOM Indexing in its...