Skip to content


Machine learning is the practice of teaching a computer to learn. The concept uses pattern recognition, as well as other forms of predictive algorithms, to make judgments on incoming data. This field is closely related to artificial intelligence and computational statistics.

Here are 2,666 public repositories matching this topic...

JonTriebenbach commented Sep 2, 2020

Bug Report

These tests were run on s390x. s390x is big-endian architecture.

Failure log for

________________________________________________ TestHelperTensorFunctions.test_make_tensor ________________________________________________

self = <helper_test.TestHelperTensorFunctions testMethod=test_make_tensor>

    def test_make_tensor(self):  # type: () -> None
PhilipMay commented Jun 12, 2020

MLflow seems to have a length limit of 5000 when setting tags (see below).

  File "/home/smay/miniconda3/envs/py38/lib/python3.8/site-packages/mlflow/utils/", line 136, in _validate_length_limit
    raise MlflowException(
mlflow.exceptions.MlflowException: Tag value '[0.8562690322984875, 0.8544098885636596, 0.8544098885636596, 0.8544098885636596, 0.85440988856365
CESARDELATORRE commented Sep 6, 2019

As mentioned by Diego, these additions would help by simplifying the API usage for users even further and it should be pretty easy to implement for us: 👍

@divega commented: dotnet/machinelearning-samples#617 (review)

@CESARDELATORRE, I did a deferred review. The experience seems pretty good.

And I agree with you that it could be even better

brunocous commented Sep 2, 2020

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 If not, is it somehow possible to define (many example for default LightGBM model) and pass a custom regression objective?

bcleenders commented Oct 2, 2020

When a user wants to stream data to a date-partitioned BQ table, the way to do this is:

//noinspection ScalaStyle
class DayPartitionFunction()
  extends SerializableFunction[ValueInSingleWindow[TableRow], TableDestination] {
  override def apply(input: ValueInSingleWindow[TableRow]): TableDestination = {
    val partition = DateTimeFormat.forPattern("yyyyMMdd").withZone(DateTimeZo
st-- commented Oct 22, 2020

Feature request

Our Scipy optimizer wrapper should explicitly handle unconnected gradients.


Currently, the GPflow Scipy optimizer wrapper (gpflow.optimizers.Scipy) simply errors (with one of the typically unhelpful TensorFlow error messages) when one of the passed-in variables is not connected to the loss to be minimized. This can easily happen in a valid case, for exa

You can’t perform that action at this time.