Configure the party yml file
Configuring the party yml file is an alternative approach for connecting parties to the aggregator.
You can get a yml file template by going into a Federated Learning experiment, clicking View setup information, and downloading the file. This is an example of a party yml file:
aggregator:
ip: [CPD_HOSTNAME]/ml/v4/trainings/[TRAINING_ID]
connection:
info:
id: [REMOTE_TRAINING_SYSTEM_ID]
"data": {
"name": "MnistSklearnDataHandler",
"path": "example.mnist_sklearn_data_handler"
"info": {
"train_file": "./mnist-keras-train.pkl",
"test_file": "./mnist-keras-test.pkl"
},
},
local_training:
name: LocalTrainingHandler
path: ibmfl.party.training.local_training_handler
protocol_handler:
name: PartyProtocolHandler
path: ibmfl.party.party_protocol_handler
Note: When the party runs the yml file from the command line, the -s
flag needs to be passed, like this command: python -m ibmfl.party.party -s <config file> <Bearer token> <log_level>
where:
<config file>
refers to the party yml file path.<log_level>
can take one of these parameters:log_level
Definition DEBUG The DEBUG log provides detailed information, typically used when diagnosing or troubleshooting problems. INFO the standard log level that indicates informative messages to ensure the software is running as expected. WARNING the log level that indicates that something unexpected happened in the application. It logs behaviors that are unexpected or are indicative of future problems but do not affect it currently. The application does not fail it continues on. ERROR the log level that must be used when the application hits an issue that prevents the functions from working. CRITICAL the log level that used to report serious errors that might prevent the continued execution of the application. <bearer_token>
: See bearer token.
Parent topic: IBM Federated Learning