Naren Castellon
07/10/2023, 8:23 PMKin Gtz. Olivares
07/10/2023, 8:28 PMNaren Castellon
07/10/2023, 8:49 PMKin Gtz. Olivares
07/10/2023, 8:51 PMimport os
os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
Naren Castellon
07/10/2023, 8:56 PMKin Gtz. Olivares
07/10/2023, 8:59 PMNaren Castellon
07/10/2023, 9:11 PMKin Gtz. Olivares
07/10/2023, 9:12 PMNaren Castellon
07/10/2023, 9:14 PMKin Gtz. Olivares
07/10/2023, 9:14 PMNaren Castellon
07/10/2023, 9:18 PMCristian (Nixtla)
07/10/2023, 9:25 PMNaren Castellon
07/10/2023, 9:28 PMCristian (Nixtla)
07/10/2023, 9:33 PMpredict
method needs to create the future values with a frequency supported by pandas, like this.Naren Castellon
07/10/2023, 9:37 PMCristian (Nixtla)
07/10/2023, 9:41 PMds
column of the future values, and nothing elseNaren Castellon
07/10/2023, 9:46 PMfede (nixtla) (they/them)
07/10/2023, 9:55 PMRMSE
? To see if there is a problem with DistributionLoss
🙌Naren Castellon
07/10/2023, 9:58 PMfede (nixtla) (they/them)
07/10/2023, 9:58 PMKMP_DUPLICATE_LIB_OK=TRUE
before launching the notebook usually solves a lot of issues with M1 🙌Naren Castellon
07/10/2023, 10:00 PMKin Gtz. Olivares
07/10/2023, 10:02 PMNaren Castellon
07/10/2023, 10:03 PMfede (nixtla) (they/them)
07/10/2023, 10:06 PMimport os
os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"