reproducible-research

With a config like this

{
    "METAFLOW_DATASTORE_SYSROOT_S3": "s3://mf-test/metaflow/",
}

(note a slash after METAFLOW_DATASTORE_SYSROOT_S3)

metaflow.S3(run=self).put* produces double-slashes like here:

s3://mf-test/metaflow//data/DataLoader/1630978962283843/month=01/data.parquet

The trailing slash in the config shouldn't make a difference

Description

Currently, when a challenge link from EvalAI is shared users see a generic view of EvalAI homepage. We want the details specific to a challenge to be shown when a link is shared. Here's how it looks currently

Expected behavior:

T

Hi @JavierAntoran @stratisMarkou,

First of all, thanks for making all of this code available - it's been great to look through!

Im currently spending some time trying to work through the Weight Uncertainty in Neural Networks in order to implement Bayes-by-Backprop. I was struggling to understand the difference between your implementation of `Bayes-by-Bac

Setting up a reader or a writer just to use them once in a straightforward way requires 5 lines of code, and it can be accomplished by using the appropriate convenience function.

itk::ReadImage and itk::WriteImage are available since ITK 5.2.0. See InsightSoftwareConsortium/ITKSphinxExamples#225
for an illustration of the changes required and https://github.com/Ins

We can wrap some import in a try ... catch statement so the user is not necessarily forced to install all the libraries, especially for the ones needed only by a specific strategy/logger.

e.g. quadprog, wandb, ...

Is your feature request related to a problem? Please describe.
For tinkering around with some of the types defined here, it could be useful to have a basic Show instance. In particular, Arg given its simplicity, and DockerTaskConfig, given that it aggregates Text and Arg values, seem not too tricky to define instances for.

Describe the solution you'd like
Data.Text already