Merge pull request #101062 from MayMSFT/patch-29

Update how-to-create-register-datasets.md

Author: PRMerger18

articles/machine-learning/how-to-create-register-datasets.md

@@ -65,21 +65,10 @@ To create datasets from an [Azure datastore](how-to-access-data.md) by using the
 
 1. Verify that you have `contributor` or `owner` access to the registered Azure datastore.
 
-1. Create the dataset by referencing a path in the datastore:
+2. Create the dataset by referencing paths in the datastore.
+> [!Note]
+> You can create a dataset from multiple paths in multiple datastores. There is no hard limit on the number of files or data size that you can create a dataset from. However, for each data path, a few requests will be sent to the storage service to check whether it points to a file or a folder. This overhead may lead to degraded performance or failure. A dataset referencing one folder with 1000 files inside is considered referencing one data path. We'd recommend creating dataset referencing less than 100 paths in datastores for optimal performance.
 
-    ```Python
-    from azureml.core.workspace import Workspace
-    from azureml.core.datastore import Datastore
-    from azureml.core.dataset import Dataset
-    
-    datastore_name = 'your datastore name'
-    
-    # get existing workspace
-    workspace = Workspace.from_config()
-    
-    # retrieve an existing datastore in the workspace by name
-    datastore = Datastore.get(workspace, datastore_name)
-    ```
 
 #### Create a TabularDataset
 
@@ -88,12 +77,20 @@ You can create TabularDatasets through the SDK or by using Azure Machine Learnin
 Use the [`from_delimited_files()`](https://docs.microsoft.com/python/api/azureml-core/azureml.data.dataset_factory.tabulardatasetfactory?view=azure-ml-py#from-delimited-files-path--validate-true--include-path-false--infer-column-types-true--set-column-types-none--separator------header-true--partition-format-none-) method on the `TabularDatasetFactory` class to read files in .csv or .tsv format, and to create an unregistered TabularDataset. If you're reading from multiple files, results will be aggregated into one tabular representation.
 
 ```Python
-# create a TabularDataset from multiple paths in datastore
-datastore_paths = [
-                  (datastore, 'weather/2018/11.csv'),
-                  (datastore, 'weather/2018/12.csv'),
-                  (datastore, 'weather/2019/*.csv')
-                 ]
+from azureml.core import Workspace, Datastore, Dataset
+
+datastore_name = 'your datastore name'
+
+# get existing workspace
+workspace = Workspace.from_config()
+    
+# retrieve an existing datastore in the workspace by name
+datastore = Datastore.get(workspace, datastore_name)
+
+# create a TabularDataset from 3 paths in datastore
+datastore_paths = [(datastore, 'ather/2018/11.csv'),
+                   (datastore, 'weather/2018/12.csv'),
+                   (datastore, 'weather/2019/*.csv')]
 weather_ds = Dataset.Tabular.from_delimited_files(path=datastore_paths)
 ```
 
@@ -154,16 +151,12 @@ Use the [`from_files()`](https://docs.microsoft.com/python/api/azureml-core/azur
 
 ```Python
 # create a FileDataset pointing to files in 'animals' folder and its subfolders recursively
-datastore_paths = [
-                  (datastore, 'animals')
-                 ]
+datastore_paths = [(datastore, 'animals')]
 animal_ds = Dataset.File.from_files(path=datastore_paths)
 
 # create a FileDataset from image and label files behind public web urls
-web_paths = [
-            'https://azureopendatastorage.blob.core.windows.net/mnist/train-images-idx3-ubyte.gz',
-            'https://azureopendatastorage.blob.core.windows.net/mnist/train-labels-idx1-ubyte.gz'
-           ]
+web_paths = ['https://azureopendatastorage.blob.core.windows.net/mnist/train-images-idx3-ubyte.gz',
+             'https://azureopendatastorage.blob.core.windows.net/mnist/train-labels-idx1-ubyte.gz']
 mnist_ds = Dataset.File.from_files(path=web_paths)
 ```
 
@@ -246,10 +239,8 @@ The dataset is now available in your workspace under **Datasets**. You can use i
 You can register a new dataset under the same name by creating a new version. A dataset version is a way to bookmark the state of your data so that you can apply a specific version of the dataset for experimentation or future reproduction. Learn more about [dataset versions](how-to-version-track-datasets.md).
 ```Python
 # create a TabularDataset from Titanic training data
-web_paths = [
-            'https://dprepdata.blob.core.windows.net/demo/Titanic.csv',
-            'https://dprepdata.blob.core.windows.net/demo/Titanic2.csv'
-           ]
+web_paths = ['https://dprepdata.blob.core.windows.net/demo/Titanic.csv',
+             'https://dprepdata.blob.core.windows.net/demo/Titanic2.csv']
 titanic_ds = Dataset.Tabular.from_delimited_files(path=web_paths)
 
 # create a new version of titanic_ds