I am struggling finding a tuturial that works on importing and using a custom dataset.

I am lost here, could someone please give me a hint ?

Even the first simple step with the simpliest dataset is not working.

Here is my simple test dataset in a csv file:

date,symbol,signal
01/01/2018,AAPL,1
02/01/2018,AAPL,0
03/01/2018,AAPL,0
04/01/2018,AAPL,0
05/01/2018,AAPL,1
06/01/2018,AAPL,0
07/01/2018,AAPL,1
08/01/2018,AAPL,0
09/01/2018,AAPL,0

Import looks "successfull"... I selected first column as primary date, and second column as primary asset :

Name Owner Type Status Created At Last Updated
test xxxx historical success 19/08/2020 à 20:01:42 19/08/2020 à 20:01:42

Then I copy your "How to use" code in a notebook :

from quantopian.pipeline import Pipeline
from quantopian.research import run_pipeline
from quantopian.pipeline.data.user_5804a659d11a8a6a1f0000d1 import test

pipe = Pipeline(
columns={
'my_dataset': test.signal.latest
},
screen=test.signal.latest.notnull()
)

df = run_pipeline(pipe, '2018-01-01', '2018-09-01')
df.head()

And obtain the following error :

ImportError Traceback (most recent call last)
in ()
1 from quantopian.pipeline import Pipeline
2 from quantopian.research import run_pipeline
----> 3 from quantopian.pipeline.data.user_5804a659d11a8a6a1f0000d1 import test
4
5 pipe = Pipeline(

ImportError: cannot import name 'test'

Which unfortunatelly doesn't tell me what the problem is.

Would be nice to have a simple csv file, say in a repo and a working notebook as an example...

What is wrong here ?

Thanks in advance for your help.