microsoft / ptgnn Goto Github PK

The provided dockerfile is not building

Hello,

Thanks for this great repo and I use it to play with varmisuse, but I have some questions about it. In the implementation, it seems that two different layers provided, create_mlp_mp_layers (see

Best wishes.

Hey! I tried to run training of the varmisuse model in order to explore how it works on data from unseen projects. I have a few questions regarding it:

1. Seems like the dataset format has changed compared to the published version of data. I've found the following issue in another repository. Unfortunately, I had already reorganized data before finding the issue: converted json files into jsonlines and changed structure from project/{train|test|valid}/files to {train|test|valid}/files. It would be nice to either duplicate the reorganizing script to this repo, or add a link to the issue in README.
2. After reorganizing the data, I tried to run training with default settings (minibatch size = 300) on an instance with 94 GB RAM and 48 CPUs. The instance doesn't have GPU because I wanted to measure the memory usage so that I can allocate a proper GPU instance afterward. Unfortunately, training fails with OOM error, because it quickly utilizes 94 GB and asks for more. Moreover, I've tried to create a smaller version of the dataset by picking only 1 project from train/validation/test, and it didn't really help: with a minibatch size of 100 and a single project in train part I still got OOM. Is it expected behavior?
3. Which instance do you recommend for training the model? In particular, how much RAM do I need and how long does the training take on, let's say, V100?
4. Do you have a pre-trained model that you can share? Maybe I can avoid the training at all and just run the already trained model on different data.

Thanks a lot in advance and thanks for great projects and papers!

Describe the new feature:

It would be great to have a conda recipe so that it can be included with projects that have more complicated build processes (for example, using libraries that need C/C++ compilers).

What is the current outcome?

Have a recipe for ptgnn on the conda-forge channel.

Is it backward-compatible?

Yes, and it would be forward-compatible as well, because the conda bots can automatically fetch new sdists pushed to pypi.

Hi,

Thanks for this wonderful work, it is really helpful for others. But when I test for the varmisuse task, it cannot run correctly. Even in the first step. May I ask for help?

Thanks

Hi Miltos,

Thanks for this great project. When I play with VarMisuse task on the released data at https://www.microsoft.com/en-us/download/details.aspx?id=56844, I face one problem.
I want to get the original function based on the graphs, however it seems that the functions are not released. So I tried to parse the built graph to restore it based on the NextToken edge but it still failed. It seems that the entrance node for the graph or the index 1 node is not the beginning of the function and the next token can not string up a completed function, see the following

The filename of this sample is 'test\Nancy.Tests\Unit\Bootstrapper\Base\BootstrapperBaseFixtureBase.cs'
So may I ask for some advice about how to get the original functions?

Thanks