-
What's the difference between pypelinin and other solutions ([gearman](http://gearman.org/index.php?id=protocol), for example)? Which one is better in which cases?
In this issue we should:
- List free…
-
Pickle memoization causes incorrect value-based hashes of dictionaries (and probably any other datastructure) containing numpy values. This snippet illustrates the problem in a reduced case:
``` pyt…
-
Hello,
I've trying to use eli5 in a MLPClassifier.
I encounter the following error:
` Error: estimator MLPClassifier(max_iter=3000, random_state=1010) is not supported `
The code that gives …
-
Related to #308 , expansion of #201
It'd be great to have a non-pickle export option. As in #308, some people want to move models to other languages, but even for those who are staying in Python, …
-
`Parallel` crashes when pickling large non-array objects under Joblib 0.10 in Python 2.7.11. What is the reason for this problem? Are there any known workarounds?
Minimalist reproduction script:
```…
-
Running pytest gtda produces two failed tests.
```
FAILED gtda/mapper/tests/test_visualization.py::test_pipeline_cloned[False] - AssertionError
FAILED gtda/mapper/tests/test_visualization.py::tes…
beew updated
4 months ago
-
Dear developers,
Recently I tried to integrate an agentpy model with the EMA workbench, the example that was on the documentation was helpful, and I was able to make that run (this example: [https:…
-
Hi! I have a question on model_T and model_Y in DML/ORF. I notice that, by default, model_T and model_Y use Lasso in scikit-learn, while I'm thinking to handle the high-dimensional confounding factors…
-
I’m unable to use num_workers>0 in Pytorch Dataloader. The reason is the low shared memory available (64M).
Can we re-direct or override the path used internally **(/dev/shm)** by Pytorch?
Exampl…
-
#### Describe the workflow you want to enable
- I would like to serialize (using `pickle` or `joblib`, as suggested in the [docs](https://scikit-learn.org/stable/modules/model_persistence.html)) a tr…