Notebook Lab

Step 1: Open Notebook

1. Log on to the PAI console, and select the target Region in the upper-left corner.

2. In the left navigation pane, choose Workspaces, then click the target workspace name to enter the workspace.

3. In the left navigation pane, choose Model Training > Data Science Workshop (DSW), then click the Notebook tab.

Step 2: Select a Code Directory

Click Select Code Directory, select an OSS bucket, and create a directory (for example, notebook_test) to use as the storage location for your Notebook files. Once configured, the Notebook list scans this directory and displays all Notebook files (*.ipynb) found under the OSS path.

Step 3: Create a Notebook and Write Code

1. Click Create Notebook. This opens the Notebook development environment and creates a new Notebook file.

2. Write your code. Click +Python to add a Python cell, then start writing code in the cell.

   The following is a sample code snippet:

   Sample code

   # Include dependency installation logic in the Notebook file so the environment
   # is set up correctly whenever you switch DSW instances.
   !pip install pandas
   # 1. Import the pandas library and alias it as pd
   import pandas as pd
   # 2. Prepare the raw data
   data = {
       'fruit': ['apple', 'banana', 'orange', 'strawberry', 'blueberry'],
       'sales_kg': [120, 200, 150, 80, 50],
       'price_per_kg': [8.5, 3.2, 5.0, 15.0, 25.0]
   }
   # 3. Create a DataFrame
   df = pd.DataFrame(data)
   # 4. Display the data
   print(df.head())

   Note

   • Install dependencies first: Because Notebook files and DSW instances are decoupled, include dependency installation commands in the Notebook file so the environment is set up correctly whenever you switch DSW instances.

   • Save your work regularly: Notebook Lab does not auto-save by default. Make sure to save your changes promptly.

Step 4: Link a DSW Instance and Run Code

1. Click Link DSW, then select an instance that is in the Running state. Multiple Notebooks can be linked to the same DSW instance, which improves resource utilization.

2. Click the icon to run the code.

   # Include dependency installation in the Notebook file to support any DSW instance.
   !pip install pandas
   import pandas as pd

   If the kernel fails to load in the Select Kernel area in the upper-right corner of the Notebook, try refreshing your browser to reload the environment.

   Sample output:

           fruit  sales_kg  price_per_kg
   0       apple       120           8.5
   1      banana       200           3.2
   2      orange       150           5.0
   3  strawberry        80          15.0
   4   blueberry        50          25.0