Programmatically set task slots per node in Azure Batch Python SDK

Hi Azure Batch Team,

I am setting up container tasks on Azure Batch using Python SDK client. I borrowed a small snippet from the GitHub, as shared here and modified it slightly to run for container tasks.

   def create_pool(batch_service_client, pool_id):  
       print('Creating pool [{}]...'.format(pool_id))  
     
       # Azure Batch Pool Image Reference  
       image_ref_to_use = batch.models.ImageReference(  
           publisher='microsoft-azure-batch',  
           offer='ubuntu-server-container',  
           sku='16-04-lts',  
           version='latest')  
     
       # VM configuration  
       vm_config_to_use=batch.models.VirtualMachineConfiguration(  
               image_reference=image_ref_to_use,  
               container_configuration=container_conf,  
               node_agent_sku_id='batch.node.ubuntu 16.04')  
     
       # Define Pool config  
       new_pool = batch.models.PoolAddParameter(  
           id=pool_id,  
           virtual_machine_configuration=vm_config_to_use,  
           vm_size='STANDARD_D1_V2',  
           target_low_priority_nodes=1,  
           task_slots_per_node=4)  
         
       batch_service_client.pool.add(new_pool)  

For the above snippet, I get a warning during pool creation that, task_slots_per_node is not a valid property of PoolAddParameter class:
task_slots_per_node is not a known attribute of class <class 'azure.batch.models._models_py3.PoolAddParameter'> and will be ignored

However, I had setup that parameter using constructor definition as suggested here. Although, the pool gets created and configuration shows task slots per node equal to 1.

So, I tried another workaround to resolve the error by setting the property after the object is created, i.e. as follows:
new_pool.task_slots_per_node=4

In this case, no warning is displayed in console. However, the task slots per node in the configuration (obtained from portal view) still shows the value equal to 1.

I would like to understand the programmatic of achieving this in Python SDK.

Regards,
Chintan Rajvir