Informacje o algorytmach & module dla projektanta Azure Machine LearningAlgorithm & module reference for Azure Machine Learning designer
Ta zawartość referencyjna zawiera informacje techniczne na temat każdego z algorytmów uczenia maszynowego i modułów dostępnych w programie Azure Machine Learning Designer.This reference content provides the technical background on each of the machine learning algorithms and modules available in Azure Machine Learning designer.
Każdy moduł reprezentuje zestaw kodu, który może zostać uruchomiony niezależnie i wykonać zadanie uczenia maszynowego z uwzględnieniem wymaganych danych wejściowych.Each module represents a set of code that can run independently and perform a machine learning task, given the required inputs. Moduł może zawierać określony algorytm lub wykonać zadanie, które jest ważne w uczeniu maszynowym, takie jak brakująca wartość zastępcza lub analiza statystyczna.A module might contain a particular algorithm, or perform a task that is important in machine learning, such as missing value replacement, or statistical analysis.
Aby uzyskać pomoc dotyczącą wybierania algorytmów, zobaczFor help with choosing algorithms, see
- Jak wybrać algorytmyHow to select algorithms
- Arkusz Ściągawka algorytmu Azure Machine LearningAzure Machine Learning Algorithm Cheat Sheet
Porada
W każdym potoku projektanta można uzyskać informacje dotyczące określonego modułu.In any pipeline in the designer, you can get information about a specific module. Wybierz link Dowiedz się więcej na karcie moduł po umieszczeniu wskaźnika na module na liście modułów lub w prawym okienku modułu.Select the Learn more link in the module card when hovering on the module in the module list, or in the right pane of the module.
Moduły przygotowywania danychData preparation modules
FunkcjaFunctionality | OpisDescription | ModułModule |
---|---|---|
Dane wejściowe i wyjścioweData Input and Output | Przenieś dane ze źródeł w chmurze do potoku.Move data from cloud sources into your pipeline. Zapisuj wyniki lub dane pośrednie w usłudze Azure Storage, SQL Database lub Hive, a następnie uruchamiaj potok lub korzystaj z magazynu w chmurze, aby wymieniać dane między potokami.Write your results or intermediate data to Azure Storage, SQL Database, or Hive, while running a pipeline, or use cloud storage to exchange data between pipelines. | Ręczne wprowadzanie danychEnter Data Manually Eksportuj daneExport Data Importuj daneImport Data |
Przekształcanie danychData Transformation | Operacje na danych, które są unikatowe dla uczenia maszynowego, takie jak normalizowanie lub pakowania danych, Zmniejszanie liczby wymiarów i konwertowanie danych między różnymi formatami plików.Operations on data that are unique to machine learning, such as normalizing or binning data, dimensionality reduction, and converting data among various file formats. | Dodaj kolumnyAdd Columns Dodawanie wierszyAdd Rows Stosowanie operacji matematycznejApply Math Operation Stosowanie przekształcenia SQLApply SQL Transformation Czyszczenie brakujących danychClean Missing Data Obcinanie wartościClip Values Konwertowanie na plik CSVConvert to CSV Konwertowanie na zestaw danychConvert to Dataset Konwertowanie na wartości wskaźnikaConvert to Indicator Values Edytowanie metadanychEdit Metadata Grupowanie danych w pojemnikiGroup Data into Bins Łączenie danychJoin Data Normalizowanie danychNormalize Data Partycjonowanie i próbkowaniePartition and Sample Usuwanie zduplikowanych wierszyRemove Duplicate Rows SMOTESMOTE Wybieranie przekształcenia kolumnSelect Columns Transform Wybieranie kolumn w zestawie danychSelect Columns in Dataset Dzielenie danychSplit Data |
Wybór funkcjiFeature Selection | Wybierz podzestaw odpowiednich, przydatnych funkcji do użycia podczas tworzenia modelu analitycznego.Select a subset of relevant, useful features to use in building an analytical model. | Wybór funkcji oparty na filtrzeFilter Based Feature Selection Ważność funkcji permutacjiPermutation Feature Importance |
Funkcje statystyczneStatistical Functions | Zapewniają szeroką gamę metod statystycznych związanych z nauką danych.Provide a wide variety of statistical methods related to data science. | Podsumowywanie danychSummarize Data |
Algorytmy uczenia maszynowegoMachine learning algorithms
Moduły do kompilowania i oceniania modeliModules for building and evaluating models
Usługa internetowaWeb service
Dowiedz się więcej na temat modułów usługi sieci Web , które są niezbędne do wnioskowania w czasie rzeczywistym w programie Azure Machine Learning Designer.Learn about the web service modules which are necessary for real-time inference in Azure Machine Learning designer.
Komunikaty o błędachError messages
Dowiedz się więcej o komunikatach o błędach i kodach wyjątków , które mogą wystąpić przy użyciu modułów w programie Azure Machine Learning Designer.Learn about the error messages and exception codes you might encounter using modules in Azure Machine Learning designer.