1
0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-14 08:16:54 +08:00

remove dataloader

This commit is contained in:
Young
2020-12-09 10:46:39 +00:00
committed by you-n-g
parent 73669de392
commit 361d168890

View File

@@ -6,7 +6,6 @@ import sys
from qlib.tests import TestAutoData from qlib.tests import TestAutoData
from qlib.data.dataset import TSDatasetH from qlib.data.dataset import TSDatasetH
import numpy as np import numpy as np
from torch.utils.data import DataLoader
import time import time
from qlib.data.dataset.handler import DataHandlerLP from qlib.data.dataset.handler import DataHandlerLP
@@ -54,21 +53,6 @@ class TestDataset(TestAutoData):
print(data.shape) print(data.shape)
print(f"2000 sample(batch index) * 20 times takes {time.time() - t}s") print(f"2000 sample(batch index) * 20 times takes {time.time() - t}s")
# FIXME: Please remove pytorch related function. Otherwise the CI tests will fail
train_loader = DataLoader(tsds_train, batch_size=800, shuffle=True, num_workers=10)
t = time.time()
for data in train_loader:
pass
print(f"Passing all training batches takes {time.time() - t}s")
# Here is an example of ffill+bfill for index
tsds_train.config(fillna_type="ffill+bfill")
train_loader = DataLoader(tsds_train, batch_size=800, shuffle=True, num_workers=10)
t = time.time()
for data in train_loader:
pass
print(f"Passing all training batches with fill takes {time.time() - t}s")
# The dimension of sample is same as tabular data, but it will return timeseries data of the sample # The dimension of sample is same as tabular data, but it will return timeseries data of the sample
# We have two method to get the time-series of a sample # We have two method to get the time-series of a sample