Examples

Note that before using the following interfaces, you can first refer to the document on-device deployment to deploy related environments.

flJobRun() for Starting Federated Learning

Before calling the flJobRun() API, instantiate the parameter class FLParameter and set related parameters as follows:

Parameter	Type	Mandatory	Description	Remarks
trainDataset	String	Y	Path of the training dataset.	The training data of a sentiment classification task is in .txt format. The training files data.bin and label.bin in the image classification task are combined by commas (,).
vocabFile	String	Y	Path of the dictionary file for data preprocessing.	This parameter is mandatory for sentiment classification tasks. This parameter does not need to be set for image classification tasks, and the default value is null.
idsFile	String	Y	Path of the mapping ID file of a dictionary.	This parameter is mandatory for sentiment classification tasks. This parameter does not need to be set for image classification tasks, and the default value is null.
testDataset	String	N	Path of the test dataset.	1. For image classification tasks, this parameter does not need to be set and the default value is null. For sentiment classification tasks, if this parameter is not set, verification is not performed during training. 2. The test data of a sentiment classification task is in .txt format. The test files data.bin and label.bin in the image classification task are combined by commas (,).
flName	String	Y	Name of the model used for federated learning.	Set this parameter to `albert` for sentiment classification tasks or `lenet` for the LeNet scenario.
trainModelPath	String	Y	Path of a training model used for federated learning, which is an absolute path of the .ms file.
inferModelPath	String	Y	Path of an inference model used for federated learning, which is an absolute path of the .ms file.	The value of this parameter need to be the same as that of trainModelPath for supervised sentiment classification tasks and image classification tasks.
useSSL	Boolean	N	Whether to perform SSL certificate authentication (which applies only to the HTTPS communication scenario) for device-cloud communication.	The value false indicates that SSL certificate authentication is not performed. The value true indicates that SSL certificate authentication is performed. The default value is false.
certPath	String	N	the SSL root certificate path used for device-cloud https communication.	This parameter must be set when the device-cloud is performing https communication and `useSSL` is set to true.
domainName	String	Y	the url for device-cloud communication.	Currently, https and http communication are supported, the corresponding formats are like: https://……, http://……, and when `useElb` is set to true, the format must be: https://127.0.0.0 : 6666 or http://127.0.0.0 : 6666 , where `127.0.0.0` corresponds to the ip of the machine providing cloud-side services (corresponding to the cloud-side parameter `--scheduler_ip`), and `6666` corresponds to the cloud-side parameter `--fl_server_port`.
ifUseElb	Boolean	Y	Whether to simulate the elastic load balancing. The value true indicates that the client sends requests to a random server address within a specified range. The value false indicates that the client sends a request to a fixed server address.	The default value is false.
serverNum	Integer	Y	Number of servers that can send requests during elastic load balancing simulation.	It can be the same as the number of servers started on the cloud when set `ifUseElb` to true, the default value is 1.

When useSSL is set to true, only HTTPS communication is supported. In addition, the following parameters need to be set:

FLParameter flParameter = FLParameter.getInstance();
String certPath  =  "client.crt";             // Provide the absolute path of the SSL root certificate.
flParameter.setCertPath(certPath);

Create a SyncFLJob object and use the flJobRun() method of the SyncFLJob class to start a federated learning task.

The sample code is as follows:

Sample code of a sentiment classification task

// set parameters
String trainDataset = "SyncFLClient/data/albert/client/0.txt";                        // Absolute path
String vocal_file = "SyncFLClient/data/albert/vocab.txt";                           // Absolute path
String idsFile = "SyncFLClient/data/albert/vocab_map_ids.txt";                      // Absolute path
String testDataset = "SyncFLClient/data/albert/eval/eval.txt";    // Absolute path. If no independent test set is included, use the training data as the test set or do not perform the test (do not set this parameter).
String flName = "albert";  
String trainModelPath = "SyncFLClient/ms/albert/albert_train.mindir.ms";                      // Absolute path
String inferModelPath = "SyncFLClient/ms/albert/albert_train.mindir.ms";                      // Absolute path
boolean useSSL = false;
String domainName = "http://10.*.*.*:6668";

FLParameter flParameter = FLParameter.getInstance();
flParameter.setTrainDataset(trainDataset);
flParameter.setVocabFile(vocabFile);
flParameter.setIdsFile(idsFile);
flParameter.setTestDataset(testDataset);
flParameter.setFlName(flName);
flParameter.setTrainModelPath(trainModelPath);
flParameter.setInferModelPath(inferModelPath);
flParameter.setUseSSL(useSSL);
flParameter.setDomainName(domainName);

// start FLJob
SyncFLJob syncFLJob = new SyncFLJob();
syncFLJob.flJobRun();

Sample code of a LeNet image classification task

// set parameters
String trainDataset = "SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_9_train_data.bin,SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_9_train_label.bin";                        // Absolute path
String testDataset = "SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_1_test_data.bin,SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_1_test_label.bin";    // Absolute path. If no independent test set is included, use the training data as the test set or do not perform the test (do not set this parameter).
String flName = "lenet";
String trainModelPath = "SyncFLClient/lenet_train.mindir0.ms";                      // Absolute path
String inferModelPath = "SyncFLClient/lenet_train.mindir0.ms";                      // Absolute path
boolean useSSL = false;
String domainName = "http://10.*.*.*:6668";

FLParameter flParameter = FLParameter.getInstance();
flParameter.setTrainDataset(trainDataset);
flParameter.setTestDataset(testDataset);
flParameter.setFlName(flName);
flParameter.setTrainModelPath(trainModelPath);
flParameter.setInferModelPath(inferModelPath);
flParameter.setUseSSL(useSSL);
flParameter.setDomainName(domainName);

// start FLJob
SyncFLJob syncFLJob = new SyncFLJob();
syncFLJob.flJobRun();

modelInference() for Inferring Multiple Input Data Records

Before calling the modelInference() API, instantiate the parameter class FLParameter and set related parameters as follows:

Parameter	Type	Mandatory	DescriptionRemarks	Remarks
flName	String	Y	Name of the model used for federated learning.	Set this parameter to `albert` for sentiment classification tasks or `lenet` for image classification tasks.
dataPath	String	Y	Dataset path.	The sentiment classification task is in .txt format without labels. The image classification task is in .bin format.
vocabFile	String	Y	Path of the dictionary file for data preprocessing.	This parameter is mandatory for sentiment classification tasks. This parameter is not needed for image classification task.
idsFile	String	Y	Path of the mapping ID file of a dictionary.	This parameter is mandatory for sentiment classification tasks. This parameter is not needed for image classification task.
modelPath	String	Y	Path of an inference model used for federated learning, which is an absolute path of the .ms file.	Both the supervised sentiment classification task and the image classification task need to be set as the same as trainModelPath that used in the federated learning training task.

Create a SyncFLJob object and use the modelInference() method of the SyncFLJob class to start an inference task on the device. The inferred label array is returned.

The sample code is as follows:

Sample code of a sentiment classification task

// set parameters
String flName = "albert";
String dataPath = "SyncFLClient/data/albert/eval/eval.txt";                             // Absolute path
String vocal_file = "SyncFLClient/data/albert/vocab.txt";                                  // Absolute path
String idsFile = "SyncFLClient/data/albert/vocab_map_ids.txt";                    // Absolute path
String modelPath = "SyncFLClient/ms/albert/albert_train.mindir.ms";                                // Absolute path
FLParameter flParameter = FLParameter.getInstance();
flParameter.setFlName(flName);
flParameter.setTestDataset(dataPath);
flParameter.setVocabFile(vocabFile);
flParameter.setIdsFile(idsFile);
flParameter.setInferModelPath(modelPath);

// inference
SyncFLJob syncFLJob = new SyncFLJob();
int[] labels = syncFLJob.modelInference();

Sample code of a LeNet image classification

// set parameters
String flName = "lenet";
String dataPath = "/SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_1_test_data.bin,/SyncFLClient/data/3500_clients_bin/f0178_39/f0178_39_bn_1_test_label.bin";     //绝对路径
String modelPath = "SyncFLClient/lenet_train.mindir0.ms";            // Absolute path
FLParameter flParameter = FLParameter.getInstance();
flParameter.setFlName(flName);
flParameter.setTestDataset(dataPath);
flParameter.setInferModelPath(modelPath);

// inference
SyncFLJob syncFLJob = new SyncFLJob();
int[] labels = syncFLJob.modelInference();

getModel() for Obtaining the Latest Model on the Cloud

Before calling the getModel() API, instantiate the parameter class FLParameter and set related parameters as follows:

Parameter	Type	Mandatory	Description	Remarks
flName	String	Y	Name of the model used for federated learning.	Set this parameter to `albert` for sentiment classification tasks or `lenet` for the LeNet scenario.
trainModelPath	String	Y	Path of a training model used for federated learning, which is an absolute path of the .ms file.
inferModelPath	String	Y	Path of an inference model used for federated learning, which is an absolute path of the .ms file.	The value of this parameter need to be the same as that of trainModelPath for supervised sentiment classification tasks and image classification tasks.
useSSL	Boolean	N	Whether to perform SSL certificate authentication (which applies only to the HTTPS communication scenario) for device-cloud communication.	The value false indicates that SSL certificate authentication is not performed. The value true indicates that SSL certificate authentication is performed. The default value is false.
certPath	String	N	the SSL root certificate path used for device-cloud https communication.	This parameter must be set when the device-cloud is performing https communication and `useSSL` is set to true.
domainName	String	Y	the url for device-cloud communication.	Currently, https and http communication are supported, the corresponding formats are like: https://……, http://……, and when `useElb` is set to true, the format must be: https://127.0.0.0 : 6666 or http://127.0.0.0 : 6666 , where `127.0.0.0` corresponds to the ip of the machine providing cloud-side services (corresponding to the cloud-side parameter `--scheduler_ip`), and `6666` corresponds to the cloud-side parameter `--fl_server_port`.
ifUseElb	Boolean	Y	Whether to simulate the elastic load balancing. The value true indicates that the client sends requests to a random server address within a specified range. The value false indicates that the client sends a request to a fixed server address.	The default value is false.
serverNum	Integer	Y	Number of servers that can send requests during elastic load balancing simulation.	It can be the same as the number of servers started on the cloud when set `ifUseElb` to true, the default value is 1.

When useSSL is set to true, only HTTPS communication is supported. In addition, the following parameters need to be set:

FLParameter flParameter = FLParameter.getInstance();
String certPath  =  "client.crt";             // Provide the absolute path of the SSL root certificate.
flParameter.setCertPath(certPath);

Create a SyncFLJob object and use the getModel() method of the SyncFLJob class to start an asynchronous inference task. The status code of the getModel request is returned.

The sample code is as follows:

Sentiment classification task

// set parameters
String flName = "albert";     // Set this parameter to `albert` for sentiment classification tasks or `lenet` for image classification tasks.
String trainModelPath = "SyncFLClient/ms/albert/albert_train.mindir.ms";                      // Absolute path
String inferModelPath = "SyncFLClient/ms/albert/albert_train.mindir.ms";                      // Absolute path
boolean useSSL = false;
String domainName = "http://10.*.*.*:6668";
boolean useElb = false;
int serverNum = 1;

FLParameter flParameter = FLParameter.getInstance();
flParameter.setFlName(flName);
flParameter.setTrainModelPath(trainModelPath);
flParameter.setInferModelPath(inferModelPath);
flParameter.setUseSSL(useSSL);
flParameter.setDomainName(domainName);
flParameter.setUseElb(useElb);
flParameter.setServerNum(serverNum);

// getModel
SyncFLJob syncFLJob = new SyncFLJob();
syncFLJob.getModel();

LeNet image classification task

// set parameters
String flName = "lenet";     // Set this parameter to `albert` for sentiment classification tasks or `lenet` for the LeNet scenario.
String trainModelPath = "SyncFLClient/lenet_train.mindir0.ms";                      // Absolute path
String inferModelPath = "SyncFLClient/lenet_train.mindir0.ms";                      // Absolute path
boolean useSSL = false;
String domainName = "http://10.*.*.*:6668";
boolean useElb = false;
int serverNum = 1;

FLParameter flParameter = FLParameter.getInstance();
flParameter.setFlName(flName);
flParameter.setTrainModelPath(trainModelPath);
flParameter.setInferModelPath(inferModelPath);
flParameter.setUseSSL(useSSL);
flParameter.setDomainName(domainName);
flParameter.setUseElb(useElb);
flParameter.setServerNum(serverNum);

// getModel
SyncFLJob syncFLJob = new SyncFLJob();
syncFLJob.getModel();