[Pytorch] DataLoader and python random module #7882
Description
Even with seeding, the following script print different ouputs for random.uniform at the different runs. random module is even reseeded here.
Outputs for torch.rand are the same though.
import torch
import random
from torch.utils.data import Dataset, DataLoader
class Data(Dataset):
def __len__(self):
return 10000
def __getitem__(self, index):
print(index, torch.rand(2, 2).sum().item(), random.uniform(0, 1))
return 1
seed = 2018
random.seed(seed)
torch.manual_seed(seed)
loader = DataLoader(Data(), num_workers=4, shuffle=True)
for x in loader:
print('-'*10)
breakFirst run
4717 2.202341079711914 0.9952153654478976
4607 2.3166141510009766 0.6813692345925851
4194 1.9806793928146362 0.6281118075687344
2595 2.95841383934021 0.8414756141240453
4691 0.9809015393257141 0.7622458327788627
9868 2.521920680999756 0.5253262288522356
7367 2.333574056625366 0.35079311205192487
9490 3.02830171585083 0.16235006783937567
----------
6759 3.1252167224884033 0.4424384676992986
Next run
4607 2.3166141510009766 0.15198273935290807
4194 1.9806793928146362 0.36414129463658884
4691 0.9809015393257141 0.027569260048619926
4717 2.202341079711914 0.5512619092026773
7367 2.333574056625366 0.7932627754589792
9490 3.02830171585083 0.19395324967791994
9868 2.521920680999756 0.5497794735158222
2595 2.95841383934021 0.782779934368899
----------
6759 3.1252167224884033 0.7098308465010348
- Ubuntu 16.04
- Python 3.6
- PyTorch version: 0.4