Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Select an option

  • Save CESARDELATORRE/38182e5f86e9df69e5dcf2f83936b2f5 to your computer and use it in GitHub Desktop.

Select an option

Save CESARDELATORRE/38182e5f86e9df69e5dcf2f83936b2f5 to your computer and use it in GitHub Desktop.

Revisions

  1. CESARDELATORRE revised this gist Jan 10, 2019. 1 changed file with 1 addition and 1 deletion.
    Original file line number Diff line number Diff line change
    @@ -1,5 +1,5 @@

    There are many more breaking changes in 0.9, but these are the ones mostly faced in the ML.NET samples:
    There are many more breaking changes in 0.9, but these are the ones that mostly the ML.NET samples faced when moving to v0.9:

    1. MakePredictionFunction() --> CreatePredictionEngine() --> Based on my original feedback and this [issue](https://github.com/dotnet/machinelearning/issues/1761), since it creates an object, not a function..

  2. CESARDELATORRE created this gist Jan 10, 2019.
    18 changes: 18 additions & 0 deletions Breaking changes from ML.NET v0.8 to v0.9 impacting ML.NET Samples
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,18 @@

    There are many more breaking changes in 0.9, but these are the ones mostly faced in the ML.NET samples:

    1. MakePredictionFunction() --> CreatePredictionEngine() --> Based on my original feedback and this [issue](https://github.com/dotnet/machinelearning/issues/1761), since it creates an object, not a function..

    2. The type or namespace name 'Microsoft.ML.Runtime' does not exist in the namespace 'Microsoft.ML' (Error: are you missing an assembly reference?)
    a. Data attributes are now in: using Microsoft.ML.Data;
    b. TextLoader and other types moved to using Microsoft.ML.Data;
    c. Related, removed:
    using Microsoft.ML.Runtime.Learners;
    using Microsoft.ML.Runtime.Data;

    3. mlContext.Data.TextReader --> mlContext.Data.CreateTextReader()
    4.
    5. RegressionEvaluator.Result -> Microsoft.ML.Data.RegressionMetrics
    6. BinaryClassifierEvaluator.Result --> Type is now Microsoft.ML.Data.CalibratedBinaryClassificationMetrics
    7. MultiClassClassifierEvaluator.Metrics --> Type is now Microsoft.ML.Data.MultiClassClassifierMetrics
    ClusteringEvaluator.Result --> Type is now ClusteringMetrics