mlops

Describe the bug
data docs columns shrink to 1 character width with long query

To Reproduce
Steps to reproduce the behavior:

1. make a batch from a long query string
2. run validation
3. render result to data docs
4. See screenshot
   <img width="1525" alt="Data_documentation_compiled_by_Great_Expectations" src="https://user-images.githubusercontent.com/928247/103230647-30eca500-4

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, one can use arithmetic operators on pipelines togher to merge them together:
Pipeline([node_a, node_b]) + Pipeline([node_c]) → Pipeline([node_a, node_b, node_c]). Currently supported operators also include -, and and or.

However, multiple users have asked that the gl

🚨🚨 Feature Request

• Related to an existing Issue
• A new implementation (Improvement, Extension)

If your feature will improve HUB

Need a way to check if a dataset already exists.

hub.empty throws an error if a dataset exists and hub.load throws an error if the dataset does not exist.

Need a way to check if a dataset already exists without throwing a

Use case

Polyaxon tracking can be used to automatically log all information generated by Ludwig.

We should probably think about adding support for Polyaxon's tracking module in Ludwig contrib.

For SC Operator it may be a good idea to generate CRD manifests from inside a docker container.
This should provide reproducible generation step and avoid "produces different output on my machine" issues.

Linter should also fail if generation of manifests produce diff with the commited version.

What steps did you take

Code gets stuck in infinite loop is SageMaker training job gets stopped (unhandled use case)

What happened:

https://github.com/kubeflow/pipelines/blob/master/components/aws/sagemaker/train/src/sagemaker_training_component.py#L57-L66

Above code only caters for training job status Completed or Failed, so if the training job status is marked as `Stopped

Describe the Issue

• Currently, we have old documentation in gh-pages but now we are not using the gh-pages for documentation and Add documentation for helm in gh-pages README
  https://github.com/flyteorg/flyte/tree/gh-pages

What if we do not do this?

Related component
Either Specific / all

• Allow tracking of a list of homogeneous objects (i.e. float values). The resulting tracked sequence is a record list, where each record is list by itself.
• Add support for getting sequence with entire lists, seqnence for the given index and for the given slice.
• Lists might have different sizes.

Example:

# track
run.track([0], ...)
run.track([1, 2, 3], ...)
run.track([4, 5, 

We're using marshmallow to parse whylogs config from YAML

However, Pydantic is much more powerful as it allows users to set config via various mechanims, from YAML, JSON to Environment settings.

We should consider moving to pydantic

Sphinx's built-in search isn't great. It'd be much better to use Algolia or similar.

Looks like it's possible:

https://stackoverflow.com/q/54872828/709975
readthedocs/sphinx_rtd_theme#761

If anyone wants to take this, I'm happy to help.