Describe the bug
When generating the batches like in experiment 27, but for the production work with the preprepared batches, all the np_batch_processors work other than the Topographic one, where in xr.combine_by_coords an error of
ValueError: Every dimension needs a coordinate for inferring concatenation order
To Reproduce
Steps to reproduce the behavior:
import glob
import os
from pathlib import Path
import numpy as np
from power_perceiver.load_prepared_batches.prepared_dataset import PreparedDataset
from power_perceiver.load_prepared_batches.data_sources import PV, GSP, HRVSatellite, NWP
from power_perceiver.production.model import FullModel
import pandas as pd
from torch.utils.data import DataLoader
dataset = PreparedDataset([PV(history_duration=pd.Timedelta("90 min"),),
GSP(history_duration=pd.Timedelta("2 hours"),),
HRVSatellite(history_duration=pd.Timedelta("30 min"),),
NWP(history_duration=pd.Timedelta("1 hour"),)],
data_path=Path("/home/jacob/Development/power_perceiver/data_for_testing/"))
print(dataset)
dataloader = DataLoader(dataset, num_workers=0, batch_size=None)
batch = next(iter(dataloader))
Expected behavior
A clear and concise description of what you expected to happen.
Additional context
Add any other context about the problem here.
Describe the bug When generating the batches like in experiment 27, but for the production work with the preprepared batches, all the np_batch_processors work other than the Topographic one, where in xr.combine_by_coords an error of
ValueError: Every dimension needs a coordinate for inferring concatenation order
To Reproduce Steps to reproduce the behavior:
Expected behavior A clear and concise description of what you expected to happen.
Additional context Add any other context about the problem here.