mirekphd

TL;DR I'd skip pinning packages with equality signs. If you must than allow for security updates by specifying a range using two inequalities: a greater or equal than the currently...

Why model.*txt* and not model.*pkl*? Have you tried the trusted `joblib.dump() / joblib.load()` pair for persisting model objects (instead of plain text)?

> do not know if and when they intend to make the vulnerability public. According to the GitHub Advisory Database PoC for both will be disclosed "on or after the...

> FYI for anyone wanting to reproduce, In my experience it is a path-dependent bug, not easy to reproduce. For instance I had the `ResolutionTooDeep` failure reproduced consistently (and stubbornly)...

I got that error too... so tried to switch over to the `keyring' package, but it did not solve the issue.

Sure. First run some python 3 env, like this container of mine: `docker run --rm -p 8888:8888 -u 1000 -v $PWD:/home/jovyan mirekphd/ml-cpu-py37-jup-cust:20202401`. Then try to run this python code in...

Adequate unbalancing is guaranteed in many containerized python applications for example, which have to be based on Ubuntu, so the base image layer is much larger than the application layers...

> Creating an empty ~/.config/containers/registries.conf makes it work. I think the root of the issue can lie in an implicit root requirement introduced by assuming that Skopeo has read access...

The solution (required for older `torch` versions, e.g. 2.0.1) was to correct the obsolete way of importing `packaging` **module `from pkg_resources` (distributed with `setuptools`** and not to be confused with...

> When running Mistral LLM with a temperature of 0, I expected the model to choose tokens with the highest probabilities exclusively. I believe that minimizing not just `temperature` *but...