Please make sure that this is a documentation issue. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:doc_template

System information

• TensorFlow version: 1.13.1
• Doc Link: https://www.tensorflow.org/api_docs/python/tf/cus

Bug Report

Is the issue related to model conversion? No

Describe the bug

DynamicQuantizeLinear function op does not have shape inference function defined. In absence of shape inference, function body is used to get the shape inference for the function op and although it works as a fallback option it hurts perf.

Expected behavior

Add shape inference function for DynamicQuan

Every kubeflow image should be scanned for security vulnerabilities.
It would be great to have a periodic security report.
Each of these images with vulnerability should be patched and updated.

URLS with the issue:

• https://mlflow.org/docs/latest/python_api/mlflow.html#mlflow.log_param
• https://mlflow.org/docs/latest/python_api/mlflow.html#mlflow.log_metric
• https://mlflow.org/docs/latest/python_api/mlflow.html#mlflow.set_tag

Description of proposal:

Document the maximum value and legal characters for log_param, log_metric and set_tag. Note that log_metric's value i

Remove logging line, or modify from ch.Info to ch.Trace:
https://github.com/dotnet/machinelearning/blob/5dbfd8acac0bf798957eea122f1413209cdf07dc/src/Microsoft.ML.Mkl.Components/SymSgdClassificationTrainer.cs#L813

For my text dataset, this logging line dumps ~100 pages of floats to my console. That level of verbosity is unneeded at the Info level.

I'd recommend just removing the loggin

Describe the Problem

plot_model currently has the save argument which can be used to save the plots. It does not provide the functionality to decide where to save the plot and with what name. Right now it saves the plot with predefined names in the current working directory.

Describe the solution you'd like

We can have another argument save_path which is used whenever the `

🚨🚨 Feature Request

We need description, citation, license, and version meta info to be added to the dataset.

Is your feature request related to a problem?

Some datasets need this info inside them for legal reasons.

If your feature will improve HUB

Easy to implement, won't hurt for sure.

Description of the possible solution

Currently, we have all metadata store

All available samples code target .Net Core, Do we have samples for .Net Framework ?

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.

Hello,

Would it be feasible to activate batching at a fixed size. As far as I know bentoML actually implement both non-batching and micro-bacthing which is adaptive one.

For certain architectures as AWS instances, the torchscript compiler is based on fixed batch size.

Thanks!

I have a simple regression task (using a LightGBMRegressor) where I want to penalize negative predictions more than positive ones. Is there a way to achieve this with the default regression LightGBM objectives (see https://lightgbm.readthedocs.io/en/latest/Parameters.html)? If not, is it somehow possible to define (many example for default LightGBM model) and pass a custom regression objective?

ideally this would demonstrate projections, predicates, & tuning options like block size (using the Configuration parameter)