Predictor

Which Predictor you use depends on how your model is exported:

• TensorFlow Predictor if your model is exported as a TensorFlow SavedModel

• Python Predictor for all other cases: PyTorch, ONNX, scikit-learn, XGBoost, TensorFlow (if not using SavedModels), etc.

The response type of the predictor can vary depending on your requirements, see HTTP API responses and gRPC API responses below.

Project files

Cortex makes all files in the project directory (i.e. the directory which contains cortex.yaml) available for use in your Predictor implementation. Python bytecode files (*.pyc, *.pyo, *.pyd), files or folders that start with ., and the api configuration file (e.g. cortex.yaml) are excluded.

The following files can also be added at the root of the project's directory:

• .cortexignore file, which follows the same syntax and behavior as a .gitignore file.

• .env file, which exports environment variables that can be used in the predictor. Each line of this file must follow the VARIABLE=value format.

For example, if your directory looks like this:

./my-classifier/
├── cortex.yaml
├── values.json
├── predictor.py
├── ...
└── requirements.txt

You can access values.json in your Predictor like this:

import json

class PythonPredictor:
    def __init__(self, config):
        with open('values.json', 'r') as values_file:
            values = json.load(values_file)
        self.values = values

HTTP

Python Predictor

Interface

# initialization code and variables can be declared here in global scope

class PythonPredictor:
    def __init__(self, config, python_client):
        """(Required) Called once before the API becomes available. Performs
        setup such as downloading/initializing the model or downloading a
        vocabulary.

        Args:
            config (required): Dictionary passed from API configuration (if
                specified). This may contain information on where to download
                the model and/or metadata.
            python_client (optional): Python client which is used to retrieve
                models for prediction. This should be saved for use in predict().
                Required when `predictor.multi_model_reloading` is specified in
                the api configuration.
        """
        self.client = python_client # optional

    def predict(self, payload, query_params, headers):
        """(Required) Called once per request. Preprocesses the request payload
        (if necessary), runs inference, and postprocesses the inference output
        (if necessary).

        Args:
            payload (optional): The request payload (see below for the possible
                payload types).
            query_params (optional): A dictionary of the query parameters used
                in the request.
            headers (optional): A dictionary of the headers sent in the request.

        Returns:
            Prediction or a batch of predictions.
        """
        pass

    def post_predict(self, response, payload, query_params, headers):
        """(Optional) Called in the background after returning a response.
        Useful for tasks that the client doesn't need to wait on before
        receiving a response such as recording metrics or storing results.

        Note: post_predict() and predict() run in the same thread pool. The
        size of the thread pool can be increased by updating
        `threads_per_process` in the api configuration yaml.

        Args:
            response (optional): The response as returned by the predict method.
            payload (optional): The request payload (see below for the possible
                payload types).
            query_params (optional): A dictionary of the query parameters used
                in the request.
            headers (optional): A dictionary of the headers sent in the request.
        """
        pass

    def load_model(self, model_path):
        """(Optional) Called by Cortex to load a model when necessary.

        This method is required when `predictor.multi_model_reloading`
        field is specified in the api configuration.

        Warning: this method must not make any modification to the model's
        contents on disk.

        Args:
            model_path: The path to the model on disk.

        Returns:
            The loaded model from disk. The returned object is what
            self.client.get_model() will return.
        """
        pass

When explicit model paths are specified in the Python predictor's API configuration, Cortex provides a python_client to your Predictor's constructor. python_client is an instance of PythonClient that is used to load model(s) (it calls the load_model() method of your predictor, which must be defined when using explicit model paths). It should be saved as an instance variable in your Predictor, and your predict() function should call python_client.get_model() to load your model for inference. Preprocessing of the JSON payload and postprocessing of predictions can be implemented in your predict() function as well.

When multiple models are defined using the Predictor's models field, the python_client.get_model() method expects an argument model_name which must hold the name of the model that you want to load (for example: self.client.get_model("text-generator")). There is also an optional second argument to specify the model version.

For proper separation of concerns, it is recommended to use the constructor's config parameter for information such as from where to download the model and initialization files, or any configurable model parameters. You define config in your API configuration, and it is passed through to your Predictor's constructor.

Your API can accept requests with different types of payloads such as JSON-parseable, bytes or starlette.datastructures.FormData data. See HTTP API requests to learn about how headers can be used to change the type of payload that is passed into your predict method.

Your predictor method can return different types of objects such as JSON-parseable, string, and bytes objects. See HTTP API responses to learn about how to configure your predictor method to respond with different response codes and content-types.

TensorFlow Predictor

Uses TensorFlow version 2.3.0 by default

Interface

class TensorFlowPredictor:
    def __init__(self, tensorflow_client, config):
        """(Required) Called once before the API becomes available. Performs
        setup such as downloading/initializing a vocabulary.

        Args:
            tensorflow_client (required): TensorFlow client which is used to
                make predictions. This should be saved for use in predict().
            config (required): Dictionary passed from API configuration (if
                specified).
        """
        self.client = tensorflow_client
        # Additional initialization may be done here

    def predict(self, payload, query_params, headers):
        """(Required) Called once per request. Preprocesses the request payload
        (if necessary), runs inference (e.g. by calling
        self.client.predict(model_input)), and postprocesses the inference
        output (if necessary).

        Args:
            payload (optional): The request payload (see below for the possible
                payload types).
            query_params (optional): A dictionary of the query parameters used
                in the request.
            headers (optional): A dictionary of the headers sent in the request.

        Returns:
            Prediction or a batch of predictions.
        """
        pass

    def post_predict(self, response, payload, query_params, headers):
        """(Optional) Called in the background after returning a response.
        Useful for tasks that the client doesn't need to wait on before
        receiving a response such as recording metrics or storing results.

        Note: post_predict() and predict() run in the same thread pool. The
        size of the thread pool can be increased by updating
        `threads_per_process` in the api configuration yaml.

        Args:
            response (optional): The response as returned by the predict method.
            payload (optional): The request payload (see below for the possible
                payload types).
            query_params (optional): A dictionary of the query parameters used
                in the request.
            headers (optional): A dictionary of the headers sent in the request.
        """
        pass

Cortex provides a tensorflow_client to your Predictor's constructor. tensorflow_client is an instance of TensorFlowClient that manages a connection to a TensorFlow Serving container to make predictions using your model. It should be saved as an instance variable in your Predictor, and your predict() function should call tensorflow_client.predict() to make an inference with your exported TensorFlow model. Preprocessing of the JSON payload and postprocessing of predictions can be implemented in your predict() function as well.

When multiple models are defined using the Predictor's models field, the tensorflow_client.predict() method expects a second argument model_name which must hold the name of the model that you want to use for inference (for example: self.client.predict(payload, "text-generator")). There is also an optional third argument to specify the model version.

For proper separation of concerns, it is recommended to use the constructor's config parameter for information such as configurable model parameters or download links for initialization files. You define config in your API configuration, and it is passed through to your Predictor's constructor.

Your API can accept requests with different types of payloads such as JSON-parseable, bytes or starlette.datastructures.FormData data. See HTTP API requests to learn about how headers can be used to change the type of payload that is passed into your predict method.

Your predictor method can return different types of objects such as JSON-parseable, string, and bytes objects. See HTTP API responses to learn about how to configure your predictor method to respond with different response codes and content-types.

If you need to share files between your predictor implementation and the TensorFlow Serving container, you can create a new directory within /mnt (e.g. /mnt/user) and write files to it. The entire /mnt directory is shared between containers, but do not write to any of the directories in /mnt that already exist (they are used internally by Cortex).

HTTP requests

The type of the payload parameter in predict(self, payload) can vary based on the content type of the request. The payload parameter is parsed according to the Content-Type header in the request. Here are the parsing rules (see below for examples):

1. For Content-Type: application/json, payload will be the parsed JSON body.

2. For Content-Type: multipart/form-data / Content-Type: application/x-www-form-urlencoded, payload will be starlette.datastructures.FormData (key-value pairs where the values are strings for text data, or starlette.datastructures.UploadFile for file uploads; see Starlette's documentation).

3. For Content-Type: text/plain, payload will be a string. utf-8 encoding is assumed, unless specified otherwise (e.g. via Content-Type: text/plain; charset=us-ascii)

4. For all other Content-Type values, payload will be the raw bytes of the request body.

Here are some examples:

JSON data

Making the request

curl http://***.amazonaws.com/my-api \
    -X POST -H "Content-Type: application/json" \
    -d '{"key": "value"}'

Reading the payload

When sending a JSON payload, the payload parameter will be a Python object:

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload):
        print(payload["key"])  # prints "value"

Binary data

Making the request

curl http://***.amazonaws.com/my-api \
    -X POST -H "Content-Type: application/octet-stream" \
    --data-binary @object.pkl

Reading the payload

Since the Content-Type: application/octet-stream header is used, the payload parameter will be a bytes object:

import pickle

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload):
        obj = pickle.loads(payload)
        print(obj["key"])  # prints "value"

Here's an example if the binary data is an image:

from PIL import Image
import io

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload, headers):
        img = Image.open(io.BytesIO(payload))  # read the payload bytes as an image
        print(img.size)

Form data (files)

Making the request

curl http://***.amazonaws.com/my-api \
    -X POST \
    -F "[email protected]" \
    -F "[email protected]" \
    -F "[email protected]"

Reading the payload

When sending files via form data, the payload parameter will be starlette.datastructures.FormData (key-value pairs where the values are starlette.datastructures.UploadFile, see Starlette's documentation). Either Content-Type: multipart/form-data or Content-Type: application/x-www-form-urlencoded can be used (typically Content-Type: multipart/form-data is used for files, and is the default in the examples above).

from PIL import Image
import pickle

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload):
        text = payload["text"].file.read()
        print(text.decode("utf-8"))  # prints the contents of text.txt

        obj = pickle.load(payload["object"].file)
        print(obj["key"])  # prints "value" assuming `object.pkl` is a pickled dictionary {"key": "value"}

        img = Image.open(payload["image"].file)
        print(img.size)  # prints the dimensions of image.png

Form data (text)

Making the request

curl http://***.amazonaws.com/my-api \
    -X POST \
    -d "key=value"

Reading the payload

When sending text via form data, the payload parameter will be starlette.datastructures.FormData (key-value pairs where the values are strings, see Starlette's documentation). Either Content-Type: multipart/form-data or Content-Type: application/x-www-form-urlencoded can be used (typically Content-Type: application/x-www-form-urlencoded is used for text, and is the default in the examples above).

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload):
        print(payload["key"])  # will print "value"

Text data

Making the request

curl http://***.amazonaws.com/my-api \
    -X POST -H "Content-Type: text/plain" \
    -d "hello world"

Reading the payload

Since the Content-Type: text/plain header is used, the payload parameter will be a string object:

class PythonPredictor:
    def __init__(self, config):
        pass

    def predict(self, payload):
        print(payload)  # prints "hello world"

HTTP responses

The response of your predict() function may be:

1. A JSON-serializable object (lists, dictionaries, numbers, etc.)

2. A string object (e.g. "class 1")

3. A bytes object (e.g. bytes(4) or pickle.dumps(obj))

4. An instance of starlette.responses.Response

gRPC

To serve your API using the gRPC protocol, make sure the predictor.protobuf_path field in your API configuration is pointing to a protobuf file. When the API gets deployed, Cortex will compile the protobuf file for its use when serving the API.

Python Predictor

Interface

# initialization code and variables can be declared here in global scope

class PythonPredictor:
    def __init__(self, config, python_client, module_proto_pb2):
        """(Required) Called once before the API becomes available. Performs
        setup such as downloading/initializing the model or downloading a
        vocabulary.

        Args:
            config (required): Dictionary passed from API configuration (if
                specified). This may contain information on where to download
                the model and/or metadata.
            python_client (optional): Python client which is used to retrieve
                models for prediction. This should be saved for use in predict().
                Required when `predictor.multi_model_reloading` is specified in
                the api configuration.
            module_proto_pb2 (optional): Loaded Python module containing the
                class definitions of the messages defined in the protobuf
                file (`predictor.protobuf_path`).
        """
        self.client = python_client # optional
        self.module_proto_pb2 = module_proto_pb2 # optional

    def predict(self, payload, context):
        """(Required) Called once per request. Preprocesses the request payload
        (if necessary), runs inference, and postprocesses the inference output
        (if necessary).

        Args:
            payload (optional): The request payload (see below for the possible
                payload types).
            context (optional): gRPC context.

        Returns:
            Prediction (when streaming is not used).

        Yield:
            Prediction (when streaming is used).
        """
        pass

    def load_model(self, model_path):
        """(Optional) Called by Cortex to load a model when necessary.

        This method is required when `predictor.multi_model_reloading`
        field is specified in the api configuration.

        Warning: this method must not make any modification to the model's
        contents on disk.

        Args:
            model_path: The path to the model on disk.

        Returns:
            The loaded model from disk. The returned object is what
            self.client.get_model() will return.
        """
        pass

When explicit model paths are specified in the Python predictor's API configuration, Cortex provides a python_client to your Predictor's constructor. python_client is an instance of PythonClient that is used to load model(s) (it calls the load_model() method of your predictor, which must be defined when using explicit model paths). It should be saved as an instance variable in your Predictor, and your predict() function should call python_client.get_model() to load your model for inference. Preprocessing of the JSON payload and postprocessing of predictions can be implemented in your predict() function as well.

When multiple models are defined using the Predictor's models field, the python_client.get_model() method expects an argument model_name which must hold the name of the model that you want to load (for example: self.client.get_model("text-generator")). There is also an optional second argument to specify the model version.

For proper separation of concerns, it is recommended to use the constructor's config parameter for information such as from where to download the model and initialization files, or any configurable model parameters. You define config in your API configuration, and it is passed through to your Predictor's constructor.

Your API can only accept the type that has been specified in the protobuf definition of your service's method. See gRPC API requests for how to construct gRPC requests.

Your predictor method can only return the type that has been specified in the protobuf definition of your service's method. See gRPC API responses for how to handle gRPC responses.

TensorFlow Predictor

Uses TensorFlow version 2.3.0 by default

Interface

class TensorFlowPredictor:
    def __init__(self, tensorflow_client, config, module_proto_pb2):
        """(Required) Called once before the API becomes available. Performs
        setup such as downloading/initializing a vocabulary.

        Args:
            tensorflow_client (required): TensorFlow client which is used to
                make predictions. This should be saved for use in predict().
            config (required): Dictionary passed from API configuration (if
                specified).
            module_proto_pb2 (optional): Loaded Python module containing the
                class definitions of the messages defined in the protobuf
                file (`predictor.protobuf_path`).
        """
        self.client = tensorflow_client
        self.module_proto_pb2 = module_proto_pb2 # optional
        # Additional initialization may be done here

    def predict(self, payload, context):
        """(Required) Called once per request. Preprocesses the request payload
        (if necessary), runs inference (e.g. by calling
        self.client.predict(model_input)), and postprocesses the inference
        output (if necessary).

        Args:
            payload (optional): The request payload (see below for the possible
                payload types).
            context (optional): gRPC context.

        Returns:
            Prediction (when streaming is not used).

        Yield:
            Prediction (when streaming is used).
        """
        pass

Cortex provides a tensorflow_client to your Predictor's constructor. tensorflow_client is an instance of TensorFlowClient that manages a connection to a TensorFlow Serving container to make predictions using your model. It should be saved as an instance variable in your Predictor, and your predict() function should call tensorflow_client.predict() to make an inference with your exported TensorFlow model. Preprocessing of the JSON payload and postprocessing of predictions can be implemented in your predict() function as well.

When multiple models are defined using the Predictor's models field, the tensorflow_client.predict() method expects a second argument model_name which must hold the name of the model that you want to use for inference (for example: self.client.predict(payload, "text-generator")). There is also an optional third argument to specify the model version.

For proper separation of concerns, it is recommended to use the constructor's config parameter for information such as configurable model parameters or download links for initialization files. You define config in your API configuration, and it is passed through to your Predictor's constructor.

Your API can only accept the type that has been specified in the protobuf definition of your service's method. See gRPC API requests for how to construct gRPC requests.

Your predictor method can only return the type that has been specified in the protobuf definition of your service's method. See gRPC API responses for how to handle gRPC responses.

If you need to share files between your predictor implementation and the TensorFlow Serving container, you can create a new directory within /mnt (e.g. /mnt/user) and write files to it. The entire /mnt directory is shared between containers, but do not write to any of the directories in /mnt that already exist (they are used internally by Cortex).

gRPC requests

Assuming the following service:

# predictor.proto

syntax = "proto3";
package sample_service;

service Predictor {
    rpc Predict (Sample) returns (Response);
}

message Sample {
    string a = 1;
}

message Response {
    string b = 1;
}

The type of the payload parameter passed into predict(self, payload) will match that of the Sample message defined in the predictor.protobuf_path file. For this example, we'll assume that the above protobuf file was specified for the API.

Simple request

The service method must look like this:

...
rpc Predict (Sample) returns (Response);
...

Making the request

import grpc, predictor_pb2, predictor_pb2_grpc

stub = predictor_pb2_grpc.PredictorStub(grpc.insecure_channel("***.amazonaws.com:80"))
stub.Predict(predictor_pb2.Sample(a="text"))

Reading the payload

In the predict method, you'll read the value like this:

...
def predict(self, payload):
    print(payload.a)
...

Streaming request

The service method must look like this:

...
rpc Predict (stream Sample) returns (Response);
...

Making the request

import grpc, predictor_pb2, predictor_pb2_grpc

def generate_iterator(sample_list):
    for sample in sample_list:
        yield sample

stub = predictor_pb2_grpc.PredictorStub(grpc.insecure_channel("***.amazonaws.com:80"))
stub.Predict(predictor_pb2.Sample(generate_iterator(["a", "b", "c", "d"])))

Reading the payload

In the predict method, you'll read the streamed values like this:

...
def predict(self, payload):
    for item in payload:
        print(item.a)
...

gRPC responses

Assuming the following service:

# predictor.proto

syntax = "proto3";
package sample_service;

service Predictor {
    rpc Predict (Sample) returns (Response);
}

message Sample {
    string a = 1;
}

message Response {
    string b = 1;
}

The type of the value that you return in your predict() method must match the Response message defined in the predictor.protobuf_path file. For this example, we'll assume that the above protobuf file was specified for the API.

Simple response

The service method must look like this:

...
rpc Predict (Sample) returns (Response);
...

Making the request

import grpc, predictor_pb2, predictor_pb2_grpc

stub = predictor_pb2_grpc.PredictorStub(grpc.insecure_channel("***.amazonaws.com:80"))
r = stub.Predict(predictor_pb2.Sample())

Returning the response

In the predict method, you'll return the value like this:

...
def predict(self, payload):
    return self.proto_module_pb2.Response(b="text")
...

Streaming response

The service method must look like this:

...
rpc Predict (Sample) returns (stream Response);
...

Making the request

import grpc, predictor_pb2, predictor_pb2_grpc

def generate_iterator(sample_list):
    for sample in sample_list:
        yield sample

stub = predictor_pb2_grpc.PredictorStub(grpc.insecure_channel("***.amazonaws.com:80"))
for r in stub.Predict(predictor_pb2.Sample())):
    print(r.b)

Returning the response

In the predict method, you'll return the streamed values like this:

...
def predict(self, payload):
    for text in ["a", "b", "c", "d"]:
        yield self.proto_module_pb2.Response(b=text)
...

Chaining APIs

It is possible to make requests from one API to another within a Cortex cluster. All running APIs are accessible from within the predictor at http://api-<api_name>:8888/predict, where <api_name> is the name of the API you are making a request to.

For example, if there is an api named text-generator running in the cluster, you could make a request to it from a different API by using:

import requests

class PythonPredictor:
    def predict(self, payload):
        response = requests.post("http://api-text-generator:8888/predict", json={"text": "machine learning is"})
        # ...

Note that the autoscaling configuration (i.e. target_replica_concurrency) for the API that is making the request should be modified with the understanding that requests will still be considered "in-flight" with the first API as the request is being fulfilled in the second API (during which it will also be considered "in-flight" with the second API).

Structured logging

You can use Cortex's logger in your predictor implemention to log in JSON. This will enrich your logs with Cortex's metadata, and you can add custom metadata to the logs by adding key value pairs to the extra key when using the logger. For example:

...
from cortex_internal.lib.log import logger as cortex_logger

class PythonPredictor:
    def predict(self, payload):
        cortex_logger.info("received payload", extra={"payload": payload})

The dictionary passed in via the extra will be flattened by one level. e.g.

{"asctime": "2021-01-19 15:14:05,291", "levelname": "INFO", "message": "received payload", "process": 235, "payload": "this movie is awesome"}

To avoid overriding essential Cortex metadata, please refrain from specifying the following extra keys: asctime, levelname, message, labels, and process. Log lines greater than 5 MB in size will be ignored.

Cortex Python client

A default Cortex Python client environment has been configured for your API. This can be used for deploying/deleting/updating or submitting jobs to your running cluster based on the execution flow of your predictor. For example:

import cortex

class PythonPredictor:
    def __init__(self, config):
        ...
        # get client pointing to the default environment
        client = cortex.client()
        # get the existing apis in the cluster for something important to you
        existing_apis = client.list_apis()