-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Need RAVEN dependency for DISPATCHES #76
Comments
@jghouse88 I'd be happy to look into this but a quick web search (e.g. https://github.com/idaholab/raven/wiki/installationLinux) seems to suggest that this could be more complicated that I would have guessed. Does anyone have more insight on which components we need and in what way are we planning to manage this as a dependency, e.g. for CI, DISPATCHES users, etc? |
@klfrick2 @joshua-cogliati-inl We can use this issue to track the dependency. Please get back here which portions of RAVEN are required for ARMA and if pip will work. Same for TEAL. |
RAVEN work on creating a wheel: idaholab/raven#1748 |
Sample pip package in: |
@joshua-cogliati-inl I tried installing Raven with pip and I got this error message:
|
@radhakrishnatg Rats. Just to check, what command did you use to get the package? |
@joshua-cogliati-inl I tried multiple commands:
and, none of them worked. All the above commands gave a similar error message. |
Which python version are you using? (At the moment, I have only built binary wheels for python 3.7) |
Ah. okay. That could be the issue. I was using python 3.10. I will switch to python 3.7 and try installing it. |
I see the same error message with python 3.7 as well
|
Okay, thank you for letting me know. I will try and figure out the problem, tho' I don't have time in the near future. |
|
After making a new raven_framework package, I was able to install raven in windows (using the miniconda python) with:
|
Thank you @joshua-cogliati-inl ! I could install raven in windows without any issues. But, I'm not sure how to use it. I tried to import raven in the following way, and it did not work:
Could you please tell me how I can run raven? |
An example input file is: https://github.com/idaholab/raven/blob/devel/doc/workshop/forwardSampling/exercises/1_sample_and_plot.xml |
@radhakrishnatg we've got some people working separately on how to use the synthetic histories via the external ROM loader as well as using TEAL for constructing pyomo objectives; as Joshua notes, RAVEN wasn't built as a library but as a code, so directly loading it as a library has limited application. |
Thank you @joshua-cogliati-inl and @PaulTalbot-INL ! I tried to run the way you suggested, and I got the following error:
As the per the suggestion, I downgraded the This time it ran with a few deprecation warnings:
But it ultimately terminated with an error message:
|
Hm, that is a strange error, (and one that I don't get from my raven pip install.). It is an error I got when I was first working on the pip package, but it should not be possible now (unless somehow an old version of Driver.py was running with a new version of Model.py) |
Oh, this was during unpickling. The pickle file must be from before the imports were fixed for the raven package (~ Mar 2022). |
@joshua-cogliati-inl Yes, I received the |
The way I have been testing is checking out raven with git, and then going to an example file (like doc/workshop/forwardSampling/exercises/1_sample_and_plot.xml ) and running raven_framework to check it. If you want a new version of something like price_sample_cluster.xml maybe @PaulTalbot-INL can point you at it. |
Thank you @joshua-cogliati-inl ! I tested with the example you mentioned, and it ran without any errors (I still see the above deprecation warnings though). |
@radhakrishnatg did Konor also provide the data files on which the We keep compatibility scripts to convert old inputs to new ones, but I don't think the |
@PaulTalbot-INL No, I did not receive the data files used for training the ARMA models. If I remember correctly, Konor used the 2019/2020 NYISO data. I don't know if this data is easily accessible or not. |
Okay. All that data should be available in https://github.com/idaholab/FORCE/tree/main/use_cases/2020_12, we can talk more about which set of data and what should be used for the use case. For now I'll focus on just the mechanics of integration and we'll work on the specifics later. |
Hm, the biggest problem with updating to newer than Python 3.7 packages is that cbcpy in pip only goes to 3.7 and that is a dependency for one of RAVEN's plugins. |
I am working on supporting Python 3.8 in: idaholab/raven#1933 |
Python 3.8 support has been added, I'll generate python 3.8 packages next time I create a new package. |
I generated new raven_framework-2.2rc3 packages in python 3.7 and 3.8 |
Thank you, @joshua-cogliati-inl I will test it either today or on Monday morning. Does this package also include the TEAL plugin? If not, could you please tell me how I can install it? |
The TEAL plugin is separate and can be installed with: |
Thank you, @joshua-cogliati-inl ! I'm seeing similar issues as before. First, it is not able to install all the required packages by itself. I had to install them manually. Second, because of the version check, I had to downgrade the version of some of the packages (
|
Hm, sometimes I have had problems with using |
Thank you @joshua-cogliati-inl Yeah, I also tried |
In python 3.8 it is now in pypi so: |
|
@lbianchi-lbl After installing Raven, I run |
RAVEN is currently not part of the required list of packages for DISPATCHES. Can we add this?
The text was updated successfully, but these errors were encountered: