Updated sample for Concatenate API. #3262
Conversation
Codecov Report
@@ Coverage Diff @@
## master #3262 +/- ##
==========================================
- Coverage 72.62% 72.62% -0.01%
==========================================
Files 807 807
Lines 145080 145080
Branches 16213 16213
==========================================
- Hits 105367 105366 -1
- Misses 35296 35297 +1
Partials 4417 4417
|
| private class InputData | ||
| { | ||
| public float Feature1; | ||
| public float[] Feature2; |
There was a problem hiding this comment.
I would put VectorType(3) here.
Otherwise you concat scalar+VarVector+Scalar which would be VarVector and you can't use it for trainers.
Or maybe include that in sample, two concats, one on varvector, one on vector.... #Resolved
|
|
||
| // A pipeline for concatenating the Age, Parity and Induced columns together into a vector that will be the Features column. | ||
| // A pipeline for concatenating the "Feature1", "Feature2" and "Feature3" columns together into a vector that will be the Features column. | ||
| // Concatenation is necessary because learners take **feature vectors** as inputs. | ||
| // e.g. var regressionTrainer = mlContext.Regression.Trainers.FastTree(labelColumn: "Label", featureColumn: "Features"); |
There was a problem hiding this comment.
is this comment necessary? #Resolved
| // | ||
| // Please note that the "Feature3" column is converted from int32 to float using the ConvertType API. | ||
| // The Concatenate API requires all the columns to be of same type. | ||
| var pipeline = mlContext.Transforms.Conversion.ConvertType("Feature4", "Feature3", outputKind: DataKind.Single) |
There was a problem hiding this comment.
"Feature4", "Feature3" [](start = 71, length = 22)
Maybe just keep calling it Feature3? #Resolved
| foreach (var featureRow in featuresColumn) | ||
| { | ||
| foreach (var value in featureRow.Features.GetValues()) | ||
| foreach (var value in featureRow.Features) | ||
| Console.Write($"{value} "); | ||
| Console.WriteLine(""); |
There was a problem hiding this comment.
How about Console.WriteLine(string.Join(" ", featureRow.Features));? #Resolved
| // 39 6 2 | ||
| // 34 4 2 | ||
| // 35 3 1 | ||
| // 0.1 1.1 2.1 3.1 1 |
There was a problem hiding this comment.
[](start = 14, length = 3)
two spaces #Resolved
|
|
||
| // A pipeline for concatenating the Age, Parity and Induced columns together into a vector that will be the Features column. | ||
| // A pipeline for concatenating the "Feature1", "Feature2" and "Feature3" columns together into a vector that will be the Features column. | ||
| // Concatenation is necessary because learners take **feature vectors** as inputs. |
There was a problem hiding this comment.
** [](start = 64, length = 2)
remove the **.
learners take vectors of floats #Resolved
There was a problem hiding this comment.
| string outputColumnName = "Features"; | ||
| var pipeline = mlContext.Transforms.Concatenate(outputColumnName, new[] { "Age", "Parity", "Induced" }); | ||
| // | ||
| // Please note that the "Feature3" column is converted from int32 to float using the ConvertType API. |
There was a problem hiding this comment.
ConvertType Transformer? #Resolved
| var pipeline = mlContext.Transforms.Concatenate(outputColumnName, new[] { "Age", "Parity", "Induced" }); | ||
| // | ||
| // Please note that the "Feature3" column is converted from int32 to float using the ConvertType API. | ||
| // The Concatenate API requires all columns to be of same type. |
There was a problem hiding this comment.
API [](start = 31, length = 3)
The Concatenate Transformer? #Resolved
ghost
locked as resolved and limited conversation to collaborators
Related to #1209.