Skip to content

Commit

Permalink
DOCS-1312: Document MLModel service API (#2111)
Browse files Browse the repository at this point in the history
  • Loading branch information
sguequierre authored Nov 2, 2023
1 parent 98ab6bc commit 73d5a57
Show file tree
Hide file tree
Showing 2 changed files with 174 additions and 0 deletions.
173 changes: 173 additions & 0 deletions docs/services/ml/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -147,6 +147,179 @@ In the absence of metadata, your `.tflite_cpu` model must satisfy the following
These requirements are satisfied by a few publicly available model architectures including EfficientDet, MobileNet, and SSD MobileNet V1.
You can use one of these architectures or build your own.

## API

The MLModel service supports the following methods:

{{< readfile "/static/include/services/apis/ml.md" >}}

{{% alert title="Tip" color="tip" %}}

The following code examples assume that you have a robot configured with an `MLModel` service, and that you add the required code to connect to your robot and import any required packages at the top of your code file.
Go to your robot's **Code Sample** tab on the [Viam app](https://app.viam.com) for boilerplate code to connect to your robot.

{{% /alert %}}

### Infer

Take an already ordered input tensor as an array, make an inference on the model, and return an output tensor map.

{{< tabs >}}
{{% tab name="Go" %}}

**Parameters:**

- `ctx` [(Context)](https://pkg.go.dev/context): A Context carries a deadline, a cancellation signal, and other values across API boundaries.
- `tensors` [(ml.Tensors)](https://pkg.go.dev/go.viam.com/[email protected]/ml#Tensors): The input map of tensors, as specified in the metadata.

**Returns:**

- [(ml.Tensors)](https://pkg.go.dev/go.viam.com/[email protected]/ml#Tensors): The output map of tensors, as specified in the metadata, after being run through an inference engine.
- [(error)](https://pkg.go.dev/builtin#error): An error, if one occurred.

For more information, see the [Go SDK Docs](https://pkg.go.dev/go.viam.com/rdk/services/mlmodel#Service).

```go {class="line-numbers linkable-line-numbers"}
myMLModel, err := mlmodel.FromRobot(robot, "my_mlmodel_service")

input_tensors := ml.Tensors{"0": tensor.New(tensor.WithShape(1, 2, 3), tensor.WithBacking(6))}

output_tensors, err := myMLModel.Infer(ctx.Background(), input_tensors)
```

{{% /tab %}}
{{% tab name="Python" %}}

**Parameters:**

- `input_tensors` [(Dict[str, NDArray])](https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html): A dictionary of input flat tensors, as specified in the metadata.
- `timeout` [(Optional\[float\])](https://docs.python.org/library/typing.html#typing.Optional): An option to set how long to wait (in seconds) before calling a time-out and closing the underlying RPC call.

**Returns:**

- [(`Dict[str, NDArray]`)](https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html): A dictionary of output flat tensors as specified in the metadata, after being run through an inference engine.

For more information, see the [Python SDK Docs](https://python.viam.dev/autoapi/viam/services/mlmodel/client/index.html#viam.services.mlmodel.client.MLModelClient.infer).

```python {class="line-numbers linkable-line-numbers"}
import numpy as np

my_mlmodel = MLModelClient.from_robot(robot=robot, name="my_mlmodel_service")

nd_array = np.array([1, 2, 3], dtype=np.float64)
input_tensors = {"0": nd_array}

output_tensors = await my_mlmodel.infer(input_tensors)
```

{{% /tab %}}
{{< /tabs >}}

### Metadata

Get the metadata: name, data type, expected tensor/array shape, inputs, and outputs associated with the ML model.

{{< tabs >}}
{{% tab name="Go" %}}

**Parameters:**

- `ctx` [(Context)](https://pkg.go.dev/context): A Context carries a deadline, a cancellation signal, and other values across API boundaries.

**Returns:**

- [(MLMetadata)](https://pkg.go.dev/go.viam.com/[email protected]/services/mlmodel#MLMetadata): Struct containing the metadata of the model file, such as the name of the model, what kind of model it is, and the expected tensor/array shape and types of the inputs and outputs of the model.
- [(error)](https://pkg.go.dev/builtin#error): An error, if one occurred.

For more information, see the [Go SDK Docs](https://pkg.go.dev/go.viam.com/rdk/services/mlmodel#Service).

```go {class="line-numbers linkable-line-numbers"}
myMLModel, err := mlmodel.FromRobot(robot, "my_mlmodel_service")

metadata, err := myMLModel.Metadata(ctx.Background())
```

{{% /tab %}}
{{% tab name="Python" %}}

**Parameters:**

- `timeout` [(Optional\[float\])](https://docs.python.org/library/typing.html#typing.Optional): An option to set how long to wait (in seconds) before calling a time-out and closing the underlying RPC call.

**Returns:**

- [(`Metadata`)](https://python.viam.dev/autoapi/viam/gen/service/mlmodel/v1/mlmodel_pb2/index.html#viam.gen.service.mlmodel.v1.mlmodel_pb2.Metadata): Name, type, expected tensor/array shape, inputs, and outputs associated with the ML model.

For more information, see the [Python SDK Docs](https://python.viam.dev/autoapi/viam/services/mlmodel/client/index.html#viam.services.mlmodel.client.MLModelClient.metadata).

```python {class="line-numbers linkable-line-numbers"}
my_mlmodel = MLModelClient.from_robot(robot=robot, name="my_mlmodel_service")

metadata = await my_mlmodel.metadata()
```

{{% /tab %}}
{{< /tabs >}}

### DoCommand

Execute model-specific commands that are not otherwise defined by the service API.
For built-in service models, any model-specific commands available are covered with each model's documentation.
If you are implementing your own navigation service and add features that have no built-in API method, you can access them with `DoCommand`.

{{< tabs >}}
{{% tab name="Go" %}}

**Parameters:**

- `ctx` [(Context)](https://pkg.go.dev/context): A Context carries a deadline, a cancellation signal, and other values across API boundaries.
- `cmd` [(map\[string\]interface{})](https://go.dev/blog/maps): The command to execute.

**Returns:**

- [(map\[string\]interface{})](https://go.dev/blog/maps): Result of the executed command.
- [(error)](https://pkg.go.dev/builtin#error): An error, if one occurred.

```go {class="line-numbers linkable-line-numbers"}
myMLModel, err := mlmodel.FromRobot(robot, "my_mlmodel_service")

resp, err := myMLModel.DoCommand(ctx, map[string]interface{}{"command": "dosomething", "someparameter": 52})
```

For more information, see the [Go SDK Docs](https://pkg.go.dev/go.viam.com/rdk/resource#Resource).

{{% /tab %}}
{{% tab name="Python" %}}

**Parameters:**

- `command` [(Mapping[str, ValueTypes])](https://docs.python.org/3/library/stdtypes.html#typesmapping): The command to execute.
- `timeout` [(Optional\[float\])](https://docs.python.org/library/typing.html#typing.Optional): An option to set how long to wait (in seconds) before calling a time-out and closing the underlying RPC call.

**Returns:**

- [(Mapping[str, ValueTypes])](https://docs.python.org/3/library/stdtypes.html#typesmapping): Result of the executed command.

**Raises:**

- `NotImplementedError`: Raised if the Resource does not support arbitrary commands.

```python {class="line-numbers linkable-line-numbers"}
my_mlmodel = MLModelClient.from_robot(robot=robot, name="my_mlmodel_service")

my_command = {
"command": "dosomething",
"someparameter": 52
}

await my_mlmodel.do_command(my_command)
```

For more information, see the [Python SDK Docs](https://python.viam.dev/autoapi/viam/services/mlmodel/client/index.html#viam.services.mlmodel.client.MLModelClient.do_command).

{{% /tab %}}
{{< /tabs >}}

## Use the ML model service with the Viam Python SDK

To use the ML model service from the [Viam Python SDK](https://python.viam.dev/), install the Python SDK using the `mlmodel` extra:
Expand Down
1 change: 1 addition & 0 deletions static/include/services/apis/ml.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,3 +2,4 @@ Method Name | Description
----------- | -----------
`Infer` | Take an already ordered input tensor as an array, make an inference on the model, and return an output tensor map.
`Metadata`| Get the metadata (such as name, type, expected tensor/array shape, inputs, and outputs) associated with the ML model.
`DoCommand` | Send arbitrary commands to the resource.

0 comments on commit 73d5a57

Please sign in to comment.