-
Hello when running:
node_brc=load(prefix+'node_brc.joblib')
edge_brc=load(prefix+'edge_brc.joblib')
I get this error
504 obj = None
505 try:
--> 506 obj = unpickler.load()
507…
-
I have trained a CNN Classifier model. I cannot save it using the .save() method. It reports an error 'CNNClassifier' object has no attribute 'model_'.
I can save it and load it using other utili…
-
Perhaps this is expected, but running the hash function on a torch Tensor gives different results with the same data. Passing the underlying numpy array does work.
Minimal example:
```python
impo…
-
I am trying to compare parallelization performance between multiprocessing and joblib and came across a strange phenomenon when combining code of both in one script:
```
import multiprocessing as mp…
-
When running a block of code from within Parallel(), it returns the error -
_pickle.PicklingError: Could not pickle the task to send it to the workers.
Is anyone facing similar issues? Any help a…
-
I'm on centos 6.3 and when I launch processes via joblib.Parallel with
the multiprocess backend, my processes all hang (inside Parallel, i
don't remember which line of code), after Parallel has launch…
ghost updated
9 years ago
-
Joblib has made it very convenient to register a new backend for execution using the `with parallel_backend(...)` context manager.
```python
with joblib.parallel_backend(...):
est.fit()
``…
-
Together with the dask team, we are considering a long-term evolution of the joblib to enable more powerful backends:
https://paper.dropbox.com/doc/Joblib-dask-integration-roadmap-ovrZdLehaltsb9qqUOE…
-
"python3 setup.py install --user" in the joblib directory install joblib as an egg. I don't understand why this is happening, but we should really avoid that. eggs are horrible because they monkey pat…
-
Joblib is printing "Joblib's exception reporting continues...". The script seems to be working fine, but I'm unsure if the print statement indicates a problem. I couldn't find anything on it in the do…