Convert ONNX neural network from fixed axes to dynamic axes and use it with ibm-watsonx-ai
¶
This notebook facilitates ONNX, PyTorch and Watson Machine Learning service. It contains steps and code to work with ibm-watsonx-ai library available in PyPI repository. It introduces commands for adapting model to dynamic axes, getting model and training data, persisting model, deploying model and scoring it.
Some familiarity with Python is helpful. This notebook uses Python 3.11.
Learning goals¶
The learning goals of this notebook are:
- Convert neural network from fixed axes to dynamic axes.
- Persist converted model in Watson Machine Learning repository.
- Deploy model for online scoring using client library.
- Score sample records using client library.
Contents¶
This notebook contains the following parts:
1. Set up the environment¶
Before you use the sample code in this notebook, you must perform the following setup tasks:
- Create a Watson Machine Learning (WML) Service instance (a free plan is offered and information about how to create the instance can be found here).
!pip install -U ibm-watsonx-ai | tail -n 1
!pip install torch==2.1 | tail -n 1
!pip install onnx==1.16 | tail -n 1
!pip install onnxruntime==1.16.3 | tail -n 1
import json
import torch
import torch.nn as nn
import onnx
import onnxruntime as ort
from onnx.tools import update_model_dims
from ibm_watsonx_ai import Credentials, APIClient
Connection to WML¶
Authenticate to the Watson Machine Learning service on IBM Cloud. You need to provide platform api_key
and instance location
.
You can use IBM Cloud CLI to retrieve platform API Key and instance location.
API Key can be generated in the following way:
ibmcloud login
ibmcloud iam api-key-create API_KEY_NAME
In result, get the value of api_key
from the output.
Location of your WML instance can be retrieved in the following way:
ibmcloud login --apikey API_KEY -a https://cloud.ibm.com
ibmcloud resource service-instance WML_INSTANCE_NAME
In result, get the value of location
from the output.
Tip: Your Cloud API key
can be generated by going to the Users section of the Cloud console. From that page, click your name, scroll down to the API Keys section, and click Create an IBM Cloud API key. Give your key a name and click Create, then copy the created key and paste it below. You can also get a service specific url by going to the Endpoint URLs section of the Watson Machine Learning docs. You can check your instance location in your Watson Machine Learning (WML) Service instance details.
You can also get service specific apikey by going to the Service IDs section of the Cloud Console. From that page, click Create, then copy the created key and paste it below.
Action: Enter your api_key
and location
in the following cell.
api_key = 'PASTE YOUR PLATFORM API KEY HERE'
location = 'PASTE YOUR INSTANCE LOCATION HERE'
credentials = Credentials(
api_key=api_key,
url=f'https://{location}.ml.cloud.ibm.com'
)
client = APIClient(credentials)
Working with spaces¶
First of all, you need to create a space that will be used for your work. If you do not have space already created, you can use Deployment Spaces Dashboard to create one.
- Click New Deployment Space
- Create an empty space
- Select Cloud Object Storage
- Select Watson Machine Learning instance and press Create
- Copy
space_id
and paste it below
Tip: You can also use SDK to prepare the space for your work. More information can be found here.
Action: Assign space ID below
space_id = 'PASTE YOUR SPACE ID HERE'
You can use list
method to print all existing spaces.
client.spaces.list(limit=10)
To be able to interact with all resources available in Watson Machine Learning, you need to set space which you will be using.
client.set.default_space(space_id)
'SUCCESS'
2. (OPTIONAL) Create and export basic ONNX model¶
This optional section demonstrates exporting a simple PyTorch model to the ONNX format. The model is configured with fixed tensor axes (i.e., dynamic_axes
are not defined), meaning it only supports a batch size of 1. The example serves as a placeholder to illustrate the process of enabling dynamic tensor axes in ONNX models later. You may replace this example model with your own or download a model that meets your specific needs.
onnx_model_path = "model.onnx"
class SimpleModel(nn.Module):
def __init__(self):
super().__init__()
self.linear = nn.Linear(2, 1)
def forward(self, x):
return self.linear(x)
model = SimpleModel()
torch_input = torch.randn(1, 2)
torch.onnx.export(model,
torch_input,
onnx_model_path,
verbose=False,
export_params=True,
keep_initializers_as_inputs=True,
input_names=['input'],
output_names=['output'])
print(f"ONNX model has been saved to {onnx_model_path}")
ONNX model has been saved to model.onnx
Initialize an ONNX Runtime session and run inference on a single batch
ort.set_default_logger_severity(3)
ort_session = ort.InferenceSession(onnx_model_path, providers=['CPUExecutionProvider'])
print("Original Input Shapes:")
for i in ort_session.get_inputs():
print(f"Input Name: {i.name}, Shape: {i.shape}")
print("\nOriginal Output Shapes:")
for o in ort_session.get_outputs():
print(f"Output Name: {o.name}, Shape: {o.shape}")
input_data = {ort_session.get_inputs()[0].name: torch_input.numpy()}
print(f"\nPrediciton result: {ort_session.run([], input_data)}")
Original Input Shapes: Input Name: input, Shape: [1, 2] Original Output Shapes: Output Name: output, Shape: [1, 1] Prediciton result: [array([[-0.39171934]], dtype=float32)]
Attempt multi-batch inference, expected to fail due to fixed axes
batch_torch_input = torch.randn(3, 2)
batch_input_data = {ort_session.get_inputs()[0].name: batch_torch_input.numpy()}
try:
batch_output = ort_session.run([], batch_input_data)
print("Multi-batch inference result:", batch_output)
except Exception as e:
print("Multi-batch inference failed as expected:", str(e))
Multi-batch inference failed as expected: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: input for the following indices index: 0 Got: 3 Expected: 1 Please fix either the inputs or the model.
3. Convert model from fixed axes to dynamic axes¶
In this section, you will convert an ONNX neural network with fixed axes to a model that supports dynamic axes. This allows for flexible input and output dimensions, making it possible to run inference with varying batch sizes.
To use your own model, provide its file path in onnx_custom_model_path
. Otherwise, the default path from the second section will be used.
onnx_custom_model_path = ''
if onnx_custom_model_path:
onnx_model_path = onnx_custom_model_path
3.1. Update Model to support dynamic batch sizes¶
You will now update the model to support dynamic axes for the first dimension (batch size) on all inputs and outputs.
model = onnx.load(onnx_model_path)
Using model.graph
here is necessary because update_inputs_outputs_dims
operates on this structure. Note that model.graph
may contain additional information beyond the primary inputs and outputs shape. Therefore, dynamic conversion is performed on the output from InferenceSession
, which accurately reflects the exact inputs and outputs.
# Initialize dictionaries to hold input and output dimensions from the model's graph
input_dims = {i.name: [d.dim_value for d in i.type.tensor_type.shape.dim] for i in model.graph.input}
output_dims = {o.name: [d.dim_value for d in o.type.tensor_type.shape.dim] for o in model.graph.output}
# Loop over all inputs/outputs and update the first axis to dynamic while keeping other axes the same
for inpt in ort_session.get_inputs():
input_dims[inpt.name] = ["batch_size"] + inpt.shape[1:]
for output in ort_session.get_outputs():
output_dims[output.name] = ["batch_size"] + output.shape[1:]
updated_model = update_model_dims.update_inputs_outputs_dims(model, input_dims, output_dims)
dynamic_model_path = "dynamic_model.onnx"
onnx.save(updated_model, dynamic_model_path)
Initialize an updated ONNX Runtime session
dynamic_ort_session = ort.InferenceSession(dynamic_model_path, providers=['CPUExecutionProvider'])
print("Original Input Shapes:")
for i in dynamic_ort_session.get_inputs():
print(f"Input Name: {i.name}, Shape: {i.shape}")
print("\nOriginal Output Shapes:")
for o in dynamic_ort_session.get_outputs():
print(f"Output Name: {o.name}, Shape: {o.shape}")
Original Input Shapes: Input Name: input, Shape: ['batch_size', 2] Original Output Shapes: Output Name: output, Shape: ['batch_size', 1]
Run inference on the updated model with a batch input
dynamic_ort_session.run([], batch_input_data)
[array([[-0.10204538], [-0.63963014], [ 0.06598502]], dtype=float32)]
The model now supports variable batch sizes, allowing for flexible batch scoring during inference.
4. Persist adjusted ONNX model¶
In this section, you will learn how to store your converted ONNX model in Watson Machine Learning repository using the IBM Watson Machine Learning SDK.
4.1. Publish model in Watson Machine Learning repository¶
Define model name, type and software spec.
sofware_spec_id = client.software_specifications.get_id_by_name("onnxruntime_opset_19")
onnx_model_zip = "onnx_dynamic_model.zip"
!zip {onnx_model_zip} {dynamic_model_path}
updating: dynamic_model.onnx (deflated 31%)
metadata = {
client.repository.ModelMetaNames.NAME: 'Dynamic axes ONNX model',
client.repository.ModelMetaNames.TYPE: 'onnxruntime_1.16',
client.repository.ModelMetaNames.SOFTWARE_SPEC_ID: sofware_spec_id
}
published_model = client.repository.store_model(
model=onnx_model_zip,
meta_props=metadata
)
4.2. Get model details¶
published_model_id = client.repository.get_model_id(published_model)
model_details = client.repository.get_details(published_model_id)
print(json.dumps(model_details, indent=2))
{ "entity": { "hybrid_pipeline_software_specs": [], "software_spec": { "id": "368d2795-aaa7-59a0-834c-248c64a5a99e", "name": "onnxruntime_opset_19" }, "type": "onnxruntime_1.16" }, "metadata": { "created_at": "2024-11-22T12:37:58.427Z", "id": "dd3a020b-315d-43bd-932e-19c3c6584dcc", "modified_at": "2024-11-22T12:38:03.717Z", "name": "Dynamic axes ONNX model", "owner": "IBMid-6980008OAA", "resource_key": "02dfbb08-a2b8-47d6-b225-2c4e1d75a72c", "space_id": "1cfa7b13-df10-4ab3-8df6-a10e8ac43780" }, "system": { "warnings": [] } }
5. Deploy and score ONNX model¶
In this section you'll learn how to create an online scoring service and predict on unseen data.
5.1. Create online deployment for published model¶
metadata = {
client.deployments.ConfigurationMetaNames.NAME: "Deployment of dynamic axes ONNX model",
client.deployments.ConfigurationMetaNames.ONLINE: {}
}
created_deployment = client.deployments.create(published_model_id, meta_props=metadata)
###################################################################################### Synchronous deployment creation for id: 'dd3a020b-315d-43bd-932e-19c3c6584dcc' started ###################################################################################### initializing Note: online_url and serving_urls are deprecated and will be removed in a future release. Use inference instead. .. ready ----------------------------------------------------------------------------------------------- Successfully finished deployment creation, deployment_id='3c9f62d4-5344-4cc7-9212-783395945540' -----------------------------------------------------------------------------------------------
deployment_id = client.deployments.get_id(created_deployment)
Now you can print an online scoring endpoint.
client.deployments.get_scoring_href(created_deployment)
You can also list existing deployments.
client.deployments.list()
5.2. Get deployment details¶
client.deployments.get_details(deployment_id)
5.3. Score¶
You can use below method to do test scoring request against deployed model.
scoring_payload = {"input_data": [{"values": batch_torch_input.tolist()}]}
Use client.deployments.score()
method to run scoring.
predictions = client.deployments.score(deployment_id, scoring_payload)
Let's print the result of predictions.
print(json.dumps(predictions, indent=2))
{ "predictions": [ { "id": "output", "values": [ [ -0.10204537957906723 ], [ -0.639630138874054 ], [ 0.06598502397537231 ] ] } ] }
As you can see, predicted values are the same one as displayed in the conversion part from test dataset.
6. Clean up¶
If you want to clean up after the notebook execution, i.e. remove any created assets like: please follow up this sample notebook.
- experiments
- trainings
- pipelines
- model definitions
- models
- functions
- deployments
7. Summary and next steps¶
You successfully completed this notebook! You learned how to use ONNX library as well as Watson Machine Learning for model conversion and deployment. Check out our Online Documentation for more samples, tutorials, documentation, how-tos, and blog posts.
Authors¶
Michał Koruszowic, Software Engineer
Copyright © 2024 IBM. This notebook and its source code are released under the terms of the MIT License.