Skip to content

Latest commit

 

History

History
321 lines (214 loc) · 10.2 KB

README.rst

File metadata and controls

321 lines (214 loc) · 10.2 KB
Raw

FATE Pipeline

Pipeline is a high-level python API that allows user to design, start, and query FATE jobs in a sequential manner. FATE Pipeline is designed to be user-friendly and consistent in behavior with FATE command line tools. User can customize job workflow by adding components to pipeline and then initiate a job with one call. In addition, Pipeline provides functionality to run prediction and query information after fitting a pipeline. Run the mini demo to have a taste of how FATE Pipeline works. Default values of party ids and work mode may need to be modified depending on the deployment setting.

python pipeline-mini-demo.py config.yaml

For more pipeline demo, please refer to examples.

A FATE Job is A Directed Acyclic Graph

A FATE job is a dag that consists of algorithm component nodes. FATE pipeline provides easy-to-use tools to configure order and setting of the tasks.

FATE is written in a modular style. Modules are designed to have input and output data and model. Therefore two modules are connected when output of one module is set to be the input of another module. By tracing how one data set is processed through FATE modules, we can see that a FATE job is in fact formed by a sequence of sub-tasks. For example, in the mini demo above, guest’s data is first read in by Reader, then loaded into DataIO. Overlapping ids between guest and host are then found by running data through Intersection. Finally, HeteroLR model is fit on the data. Each listed modules run a small task with the data, and together they constitute a model training job.

Beyond the given mini demo, a job may include multiple data sets and models. For more pipeline examples, please refer to examples.

Install Pipeline

Pipeline CLI

After successfully installed FATE Client, user needs to configure server information and log directory for Pipeline. Pipeline provides a command line tool for quick setup. Run the following command for more information.

pipeline init --help

Interface of Pipeline

Component

FATE modules are wrapped into component in Pipeline API. Each component can take in and output Data and Model. Parameters of components can be set conveniently at the time of initialization. Unspecified parameters will take default values. All components have a name, which can be arbitrarily set. A component’s name is its identifier, and so it must be unique within a pipeline. We suggest that each component name includes a numbering as suffix for easy tracking.

Components each may have input and/or output Data and/or Model. For details on how to use component, please refer to this guide.

An example of initializing a component with specified parameter values:

hetero_lr_0 = HeteroLR(name="hetero_lr_0", early_stop="weight_diff", max_iter=10,
                       early_stopping_rounds=2, validation_freqs=2)

Input

Input encapsulates all input of a component, including Data and Model input. To access input of a component, reference its input attribute:

input_all = dataio_0.input

Output

Output encapsulates all output result of a component, including Data and Model output. To access Output from a component, reference its output attribute:

output_all = dataio_0.output

Data

Data wraps all data-type input and output of components. FATE Pipeline includes five types of data, each is used for different scenario. For more information, please refer here.

Model

Model defines model input and output of components. Similar to Data, the two types of models are used for different purposes. For more information, please refer here.

Build A Pipeline

Below is a general guide to building a pipeline. Please refer to mini demo for an explained demo.

Once initialized a pipeline, job participants and initiator should be specified. Below is an example of initial setup of a pipeline:

pipeline = PipeLine()
pipeline.set_initiator(role='guest', party_id=9999)
pipeline.set_roles(guest=9999, host=10000, arbiter=10000)

Reader is required to read in data source so that other component(s) can process data. Define a Reader component:

reader_0 = Reader(name="reader_0")

In most cases, DataIO follows Reader to transform data into DataInstance format, which can then be used for data engineering and model training. Some components (such as Union and Intersection) can run directly on non-DataInstance tables.

All pipeline components can be configured individually for different roles by setting get_party_instance. For instance, DataIO component can be configured specifically for guest like this:

dataio_0 = DataIO(name="dataio_0")
guest_component_instance = dataio_0.get_party_instance(role='guest', party_id=9999)
guest_component_instance.component_param(with_label=True, output_format="dense")

To include a component in a pipeline, use add_component. To add the DataIO component to the previously created pipeline, try this:

pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))

Build Fate NN Model In Keras Style

In pipeline, you can build NN structures in a Keras style. Take Homo-NN as an example:

First, import Keras and define your nn structures:

from tensorflow.keras import optimizers
from tensorflow.keras.layers import Dense

layer_0 = Dense(units=6, input_shape=(10,), activation="relu")
layer_1 = Dense(units=1, activation="sigmoid")

Then, add nn layers into Homo-NN model like using Sequential class in Keras:

from pipeline.component.homo_nn import HomoNN

# set parameter
homo_nn_0 = HomoNN(name="homo_nn_0", max_iter=10, batch_size=-1, early_stop={"early_stop": "diff", "eps": 0.0001})
homo_nn_0.add(layer_0)
homo_nn_0.add(layer_1)

Set optimizer and compile Homo-NN model:

homo_nn_0.compile(optimizer=optimizers.Adam(learning_rate=0.05), metrics=["Hinge", "accuracy", "AUC"],
                  loss="binary_crossentropy")

Add it to pipeline:

pipeline.add_component(homo_nn, data=Data(train_data=dataio_0.output.data))

Init Runtime JobParameters

To fit or predict, user needs to initialize the runtime environment, like 'backend' and 'work_mode',

from pipeline.runtime.entity import JobParameters
job_parameters = JobParameters(backend=Backend.EGGROLL, work_mode=WorkMode.STANDALONE)

Run A Pipeline

Having added all components, user needs to first compile pipeline before running the designed job. After compilation, the pipeline can then be fit(run train job) with appropriate Backend and WorkMode.

pipeline.compile()
pipeline.fit(job_parameters)

Query on Tasks

FATE Pipeline provides API to query component information, including data, model, and summary. All query API have matching name to FlowPy, while Pipeline retrieves and returns query result directly to user.

summary = pipeline.get_component("hetero_lr_0").get_summary()

Deploy Components

Once fitting pipeline completes, prediction can be run on new data set. Before prediction, necessary components need to be first deployed. This step marks selected components to be used by prediction pipeline.

# deploy select components
pipeline.deploy_component([dataio_0, hetero_lr_0])
# deploy all components
# note that Reader component cannot be deployed. Always deploy pipeline with Reader by specified component list.
pipeline.deploy_component()

Predict with Pipeline

First, initiate a new pipeline, then specify data source used for prediction.

predict_pipeline = PipeLine()
predict_pipeline.add_component(reader_0)
predict_pipeline.add_component(pipeline,
                               data=Data(predict_input={pipeline.dataio_0.input.data: reader_0.output.data}))

Prediction can then be initiated on the new pipeline.

predict_pipeline.predict(job_parameters)

In addition, since pipeline is modular, user may add new components to the original pipeline before running prediction.

predict_pipeline.add_component(evaluation_0, data=Data(data=pipeline.hetero_lr_0.output.data))
predict_pipeline.predict(job_parameters)

Save and Recovery of Pipeline

To save a pipeline, just use dump interface.

pipeline.dump("pipeline_saved.pkl")

To restore a pipeline, use load_model_from_file interface.

from pipeline.backend.pipeline import PineLine
PipeLine.load_model_from_file("pipeline_saved.pkl")

Summary Info of Pipeline

To get the details of a pipeline, use describe interface, which prints the "create time" fit or predict state and the constructed dsl if exists.

pipeline.describe()

Upload Data

Pipeline provides functionality to upload local data table. Please refer to upload demo for a quick example. Note that uploading data can be added all at once, and the pipeline used to perform upload can be either training or prediction pipeline (or, a separate pipeline as in the demo).

Pipeline vs. CLI

In the past versions, user interacts with FATE through command line interface, often with manually configured conf and dsl json files. Manual configuration can be tedious and error-prone. FATE Pipeline forms task configure files automatically at compilation, allowing quick experiment with task design.