Reputation: 471
Using Pycharm Community 2018.1.4
Python 3.6
Dask 2.8.1
Trying to implement dask delayed on some of my methods and getting an error
AttributeError: module 'dask' has no attribute 'delayed'.
This is obviously not true so I am wondering what I am doing wrong. My implementation structure is as follows:
import dask
def main()
for i, fn in enumarate(filenames):
data = {}
for x in range(0,2):
data.update(dask.delayed(load_data)(fn, x))
succes_flag = dask.delayed(execute_analytic)(data)
if success_flag == 1:
print("success")
else:
print("fail")
def load_data(filename,selector):
def execute_analytic(data)
if __name__ == '__main__':
dask.compute(main())
Essentialy, I have a bunch of data files, which are independant of each other and so I want to run them in parallel instead of sequentially through a for loop, which i was doing if you take the dask.delayed out.
Am i fundamentally missing anything in the above implementation of dask delayed?
Upvotes: 3
Views: 10645
Reputation: 14769
pip install "dask[delayed]"
is the minimal requirement to directly answer the OP (the other answers may install unnecessary dependencies)
Upvotes: 2
Reputation: 310
I refer the following URL https://github.com/dask/dask/issues/1849
To install Dask with pip there are a few options, depending on which dependencies you would like to keep up to date:
Upvotes: 6
Reputation: 57251
You probably only installed the core library, rather than the full library with normal dependencies.
conda install dask
or
pip install dask[complete]
See https://docs.dask.org/en/latest/install.html for more information
Upvotes: 3