flwr-nightly 1.8.0.dev20240315__py3-none-any.whl → 1.15.0.dev20250115__py3-none-any.whl

flwr/server/strategy/qfedavg.py

@@ -1,4 +1,4 @@
-# Copyright 2020 Flower Labs GmbH. All Rights Reserved.
+# Copyright 2021 Flower Labs GmbH. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -19,7 +19,7 @@ Paper: openreview.net/pdf?id=ByexElSYDr
 
 
 from logging import WARNING
-from typing import Callable, Dict, List, Optional, Tuple, Union
+from typing import Callable, Optional, Union
 
 import numpy as np
 
@@ -60,12 +60,12 @@ class QFedAvg(FedAvg):
         min_available_clients: int = 1,
         evaluate_fn: Optional[
             Callable[
-                [int, NDArrays, Dict[str, Scalar]],
-                Optional[Tuple[float, Dict[str, Scalar]]],
+                [int, NDArrays, dict[str, Scalar]],
+                Optional[tuple[float, dict[str, Scalar]]],
             ]
         ] = None,
-        on_fit_config_fn: Optional[Callable[[int], Dict[str, Scalar]]] = None,
-        on_evaluate_config_fn: Optional[Callable[[int], Dict[str, Scalar]]] = None,
+        on_fit_config_fn: Optional[Callable[[int], dict[str, Scalar]]] = None,
+        on_evaluate_config_fn: Optional[Callable[[int], dict[str, Scalar]]] = None,
         accept_failures: bool = True,
         initial_parameters: Optional[Parameters] = None,
         fit_metrics_aggregation_fn: Optional[MetricsAggregationFn] = None,
@@ -95,19 +95,19 @@ class QFedAvg(FedAvg):
         rep += f"q_param={self.q_param}, pre_weights={self.pre_weights})"
         return rep
 
-    def num_fit_clients(self, num_available_clients: int) -> Tuple[int, int]:
+    def num_fit_clients(self, num_available_clients: int) -> tuple[int, int]:
         """Return the sample size and the required number of available clients."""
         num_clients = int(num_available_clients * self.fraction_fit)
         return max(num_clients, self.min_fit_clients), self.min_available_clients
 
-    def num_evaluation_clients(self, num_available_clients: int) -> Tuple[int, int]:
+    def num_evaluation_clients(self, num_available_clients: int) -> tuple[int, int]:
         """Use a fraction of available clients for evaluation."""
         num_clients = int(num_available_clients * self.fraction_evaluate)
         return max(num_clients, self.min_evaluate_clients), self.min_available_clients
 
     def configure_fit(
         self, server_round: int, parameters: Parameters, client_manager: ClientManager
-    ) -> List[Tuple[ClientProxy, FitIns]]:
+    ) -> list[tuple[ClientProxy, FitIns]]:
         """Configure the next round of training."""
         weights = parameters_to_ndarrays(parameters)
         self.pre_weights = weights
@@ -131,7 +131,7 @@ class QFedAvg(FedAvg):
 
     def configure_evaluate(
         self, server_round: int, parameters: Parameters, client_manager: ClientManager
-    ) -> List[Tuple[ClientProxy, EvaluateIns]]:
+    ) -> list[tuple[ClientProxy, EvaluateIns]]:
         """Configure the next round of evaluation."""
         # Do not configure federated evaluation if fraction_evaluate is 0
         if self.fraction_evaluate == 0.0:
@@ -158,9 +158,9 @@ class QFedAvg(FedAvg):
     def aggregate_fit(
         self,
         server_round: int,
-        results: List[Tuple[ClientProxy, FitRes]],
-        failures: List[Union[Tuple[ClientProxy, FitRes], BaseException]],
-    ) -> Tuple[Optional[Parameters], Dict[str, Scalar]]:
+        results: list[tuple[ClientProxy, FitRes]],
+        failures: list[Union[tuple[ClientProxy, FitRes], BaseException]],
+    ) -> tuple[Optional[Parameters], dict[str, Scalar]]:
         """Aggregate fit results using weighted average."""
         if not results:
             return None, {}
@@ -229,9 +229,9 @@ class QFedAvg(FedAvg):
     def aggregate_evaluate(
         self,
         server_round: int,
-        results: List[Tuple[ClientProxy, EvaluateRes]],
-        failures: List[Union[Tuple[ClientProxy, EvaluateRes], BaseException]],
-    ) -> Tuple[Optional[float], Dict[str, Scalar]]:
+        results: list[tuple[ClientProxy, EvaluateRes]],
+        failures: list[Union[tuple[ClientProxy, EvaluateRes], BaseException]],
+    ) -> tuple[Optional[float], dict[str, Scalar]]:
         """Aggregate evaluation losses using weighted average."""
         if not results:
             return None, {}

flwr/server/strategy/strategy.py

@@ -16,7 +16,7 @@
 
 
 from abc import ABC, abstractmethod
-from typing import Dict, List, Optional, Tuple, Union
+from typing import Optional, Union
 
 from flwr.common import EvaluateIns, EvaluateRes, FitIns, FitRes, Parameters, Scalar
 from flwr.server.client_manager import ClientManager
@@ -47,7 +47,7 @@ class Strategy(ABC):
     @abstractmethod
     def configure_fit(
         self, server_round: int, parameters: Parameters, client_manager: ClientManager
-    ) -> List[Tuple[ClientProxy, FitIns]]:
+    ) -> list[tuple[ClientProxy, FitIns]]:
         """Configure the next round of training.
 
         Parameters
@@ -72,9 +72,9 @@ class Strategy(ABC):
     def aggregate_fit(
         self,
         server_round: int,
-        results: List[Tuple[ClientProxy, FitRes]],
-        failures: List[Union[Tuple[ClientProxy, FitRes], BaseException]],
-    ) -> Tuple[Optional[Parameters], Dict[str, Scalar]]:
+        results: list[tuple[ClientProxy, FitRes]],
+        failures: list[Union[tuple[ClientProxy, FitRes], BaseException]],
+    ) -> tuple[Optional[Parameters], dict[str, Scalar]]:
         """Aggregate training results.
 
         Parameters
@@ -108,7 +108,7 @@ class Strategy(ABC):
     @abstractmethod
     def configure_evaluate(
         self, server_round: int, parameters: Parameters, client_manager: ClientManager
-    ) -> List[Tuple[ClientProxy, EvaluateIns]]:
+    ) -> list[tuple[ClientProxy, EvaluateIns]]:
         """Configure the next round of evaluation.
 
         Parameters
@@ -134,9 +134,9 @@ class Strategy(ABC):
     def aggregate_evaluate(
         self,
         server_round: int,
-        results: List[Tuple[ClientProxy, EvaluateRes]],
-        failures: List[Union[Tuple[ClientProxy, EvaluateRes], BaseException]],
-    ) -> Tuple[Optional[float], Dict[str, Scalar]]:
+        results: list[tuple[ClientProxy, EvaluateRes]],
+        failures: list[Union[tuple[ClientProxy, EvaluateRes], BaseException]],
+    ) -> tuple[Optional[float], dict[str, Scalar]]:
         """Aggregate evaluation results.
 
         Parameters
@@ -164,7 +164,7 @@ class Strategy(ABC):
     @abstractmethod
     def evaluate(
         self, server_round: int, parameters: Parameters
-    ) -> Optional[Tuple[float, Dict[str, Scalar]]]:
+    ) -> Optional[tuple[float, dict[str, Scalar]]]:
         """Evaluate the current model parameters.
 
         This function can be used to perform centralized (i.e., server-side) evaluation

flwr/server/superlink/driver/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2022 Flower Labs GmbH. All Rights Reserved.
+# Copyright 2024 Flower Labs GmbH. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 # ==============================================================================
-"""Flower driver service."""
+"""Flower ServerAppIo service."""

flwr/server/superlink/driver/serverappio_grpc.py

@@ -0,0 +1,61 @@
+# Copyright 2024 Flower Labs GmbH. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""ServerAppIo gRPC API."""
+
+
+from logging import INFO
+from typing import Optional
+
+import grpc
+
+from flwr.common import GRPC_MAX_MESSAGE_LENGTH
+from flwr.common.grpc import generic_create_grpc_server
+from flwr.common.logger import log
+from flwr.proto.serverappio_pb2_grpc import (  # pylint: disable=E0611
+    add_ServerAppIoServicer_to_server,
+)
+from flwr.server.superlink.ffs.ffs_factory import FfsFactory
+from flwr.server.superlink.linkstate import LinkStateFactory
+
+from .serverappio_servicer import ServerAppIoServicer
+
+
+def run_serverappio_api_grpc(
+    address: str,
+    state_factory: LinkStateFactory,
+    ffs_factory: FfsFactory,
+    certificates: Optional[tuple[bytes, bytes, bytes]],
+) -> grpc.Server:
+    """Run ServerAppIo API (gRPC, request-response)."""
+    # Create ServerAppIo API gRPC server
+    serverappio_servicer: grpc.Server = ServerAppIoServicer(
+        state_factory=state_factory,
+        ffs_factory=ffs_factory,
+    )
+    serverappio_add_servicer_to_server_fn = add_ServerAppIoServicer_to_server
+    serverappio_grpc_server = generic_create_grpc_server(
+        servicer_and_add_fn=(
+            serverappio_servicer,
+            serverappio_add_servicer_to_server_fn,
+        ),
+        server_address=address,
+        max_message_length=GRPC_MAX_MESSAGE_LENGTH,
+        certificates=certificates,
+    )
+
+    log(INFO, "Flower ECE: Starting ServerAppIo API (gRPC-rere) on %s", address)
+    serverappio_grpc_server.start()
+
+    return serverappio_grpc_server

flwr/server/superlink/driver/serverappio_servicer.py

@@ -0,0 +1,361 @@
+# Copyright 2024 Flower Labs GmbH. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""ServerAppIo API servicer."""
+
+
+import threading
+import time
+from logging import DEBUG, INFO
+from typing import Optional
+from uuid import UUID
+
+import grpc
+
+from flwr.common import ConfigsRecord
+from flwr.common.constant import Status
+from flwr.common.logger import log
+from flwr.common.serde import (
+    context_from_proto,
+    context_to_proto,
+    fab_from_proto,
+    fab_to_proto,
+    run_status_from_proto,
+    run_status_to_proto,
+    run_to_proto,
+    user_config_from_proto,
+)
+from flwr.common.typing import Fab, RunStatus
+from flwr.proto import serverappio_pb2_grpc  # pylint: disable=E0611
+from flwr.proto.fab_pb2 import GetFabRequest, GetFabResponse  # pylint: disable=E0611
+from flwr.proto.log_pb2 import (  # pylint: disable=E0611
+    PushLogsRequest,
+    PushLogsResponse,
+)
+from flwr.proto.node_pb2 import Node  # pylint: disable=E0611
+from flwr.proto.run_pb2 import (  # pylint: disable=E0611
+    CreateRunRequest,
+    CreateRunResponse,
+    GetRunRequest,
+    GetRunResponse,
+    GetRunStatusRequest,
+    GetRunStatusResponse,
+    UpdateRunStatusRequest,
+    UpdateRunStatusResponse,
+)
+from flwr.proto.serverappio_pb2 import (  # pylint: disable=E0611
+    GetNodesRequest,
+    GetNodesResponse,
+    PullServerAppInputsRequest,
+    PullServerAppInputsResponse,
+    PullTaskResRequest,
+    PullTaskResResponse,
+    PushServerAppOutputsRequest,
+    PushServerAppOutputsResponse,
+    PushTaskInsRequest,
+    PushTaskInsResponse,
+)
+from flwr.proto.task_pb2 import TaskRes  # pylint: disable=E0611
+from flwr.server.superlink.ffs.ffs import Ffs
+from flwr.server.superlink.ffs.ffs_factory import FfsFactory
+from flwr.server.superlink.linkstate import LinkState, LinkStateFactory
+from flwr.server.superlink.utils import abort_if
+from flwr.server.utils.validator import validate_task_ins_or_res
+
+
+class ServerAppIoServicer(serverappio_pb2_grpc.ServerAppIoServicer):
+    """ServerAppIo API servicer."""
+
+    def __init__(
+        self, state_factory: LinkStateFactory, ffs_factory: FfsFactory
+    ) -> None:
+        self.state_factory = state_factory
+        self.ffs_factory = ffs_factory
+        self.lock = threading.RLock()
+
+    def GetNodes(
+        self, request: GetNodesRequest, context: grpc.ServicerContext
+    ) -> GetNodesResponse:
+        """Get available nodes."""
+        log(DEBUG, "ServerAppIoServicer.GetNodes")
+
+        # Init state
+        state: LinkState = self.state_factory.state()
+
+        # Abort if the run is not running
+        abort_if(
+            request.run_id,
+            [Status.PENDING, Status.STARTING, Status.FINISHED],
+            state,
+            context,
+        )
+
+        all_ids: set[int] = state.get_nodes(request.run_id)
+        nodes: list[Node] = [Node(node_id=node_id) for node_id in all_ids]
+        return GetNodesResponse(nodes=nodes)
+
+    def CreateRun(
+        self, request: CreateRunRequest, context: grpc.ServicerContext
+    ) -> CreateRunResponse:
+        """Create run ID."""
+        log(DEBUG, "ServerAppIoServicer.CreateRun")
+        state: LinkState = self.state_factory.state()
+        if request.HasField("fab"):
+            fab = fab_from_proto(request.fab)
+            ffs: Ffs = self.ffs_factory.ffs()
+            fab_hash = ffs.put(fab.content, {})
+            _raise_if(
+                validation_error=fab_hash != fab.hash_str,
+                request_name="CreateRun",
+                detail=f"FAB ({fab.hash_str}) hash from request doesn't match contents",
+            )
+        else:
+            fab_hash = ""
+        run_id = state.create_run(
+            request.fab_id,
+            request.fab_version,
+            fab_hash,
+            user_config_from_proto(request.override_config),
+            ConfigsRecord(),
+        )
+        return CreateRunResponse(run_id=run_id)
+
+    def PushTaskIns(
+        self, request: PushTaskInsRequest, context: grpc.ServicerContext
+    ) -> PushTaskInsResponse:
+        """Push a set of TaskIns."""
+        log(DEBUG, "ServerAppIoServicer.PushTaskIns")
+
+        # Init state
+        state: LinkState = self.state_factory.state()
+
+        # Abort if the run is not running
+        abort_if(
+            request.run_id,
+            [Status.PENDING, Status.STARTING, Status.FINISHED],
+            state,
+            context,
+        )
+
+        # Set pushed_at (timestamp in seconds)
+        pushed_at = time.time()
+        for task_ins in request.task_ins_list:
+            task_ins.task.pushed_at = pushed_at
+
+        # Validate request
+        _raise_if(
+            validation_error=len(request.task_ins_list) == 0,
+            request_name="PushTaskIns",
+            detail="`task_ins_list` must not be empty",
+        )
+        for task_ins in request.task_ins_list:
+            validation_errors = validate_task_ins_or_res(task_ins)
+            _raise_if(
+                validation_error=bool(validation_errors),
+                request_name="PushTaskIns",
+                detail=", ".join(validation_errors),
+            )
+            _raise_if(
+                validation_error=request.run_id != task_ins.run_id,
+                request_name="PushTaskIns",
+                detail="`task_ins` has mismatched `run_id`",
+            )
+
+        # Store each TaskIns
+        task_ids: list[Optional[UUID]] = []
+        for task_ins in request.task_ins_list:
+            task_id: Optional[UUID] = state.store_task_ins(task_ins=task_ins)
+            task_ids.append(task_id)
+
+        return PushTaskInsResponse(
+            task_ids=[str(task_id) if task_id else "" for task_id in task_ids]
+        )
+
+    def PullTaskRes(
+        self, request: PullTaskResRequest, context: grpc.ServicerContext
+    ) -> PullTaskResResponse:
+        """Pull a set of TaskRes."""
+        log(DEBUG, "ServerAppIoServicer.PullTaskRes")
+
+        # Init state
+        state: LinkState = self.state_factory.state()
+
+        # Abort if the run is not running
+        abort_if(
+            request.run_id,
+            [Status.PENDING, Status.STARTING, Status.FINISHED],
+            state,
+            context,
+        )
+
+        # Convert each task_id str to UUID
+        task_ids: set[UUID] = {UUID(task_id) for task_id in request.task_ids}
+
+        # Read from state
+        task_res_list: list[TaskRes] = state.get_task_res(task_ids=task_ids)
+
+        # Validate request
+        for task_res in task_res_list:
+            _raise_if(
+                validation_error=request.run_id != task_res.run_id,
+                request_name="PullTaskRes",
+                detail="`task_res` has mismatched `run_id`",
+            )
+
+        # Delete the TaskIns/TaskRes pairs if TaskRes is found
+        task_ins_ids_to_delete = {
+            UUID(task_res.task.ancestry[0]) for task_res in task_res_list
+        }
+        state.delete_tasks(task_ins_ids=task_ins_ids_to_delete)
+
+        return PullTaskResResponse(task_res_list=task_res_list)
+
+    def GetRun(
+        self, request: GetRunRequest, context: grpc.ServicerContext
+    ) -> GetRunResponse:
+        """Get run information."""
+        log(DEBUG, "ServerAppIoServicer.GetRun")
+
+        # Init state
+        state: LinkState = self.state_factory.state()
+
+        # Retrieve run information
+        run = state.get_run(request.run_id)
+
+        if run is None:
+            return GetRunResponse()
+
+        return GetRunResponse(run=run_to_proto(run))
+
+    def GetFab(
+        self, request: GetFabRequest, context: grpc.ServicerContext
+    ) -> GetFabResponse:
+        """Get FAB from Ffs."""
+        log(DEBUG, "ServerAppIoServicer.GetFab")
+
+        ffs: Ffs = self.ffs_factory.ffs()
+        if result := ffs.get(request.hash_str):
+            fab = Fab(request.hash_str, result[0])
+            return GetFabResponse(fab=fab_to_proto(fab))
+
+        raise ValueError(f"Found no FAB with hash: {request.hash_str}")
+
+    def PullServerAppInputs(
+        self, request: PullServerAppInputsRequest, context: grpc.ServicerContext
+    ) -> PullServerAppInputsResponse:
+        """Pull ServerApp process inputs."""
+        log(DEBUG, "ServerAppIoServicer.PullServerAppInputs")
+        # Init access to LinkState and Ffs
+        state = self.state_factory.state()
+        ffs = self.ffs_factory.ffs()
+
+        # Lock access to LinkState, preventing obtaining the same pending run_id
+        with self.lock:
+            # Attempt getting the run_id of a pending run
+            run_id = state.get_pending_run_id()
+            # If there's no pending run, return an empty response
+            if run_id is None:
+                return PullServerAppInputsResponse()
+
+            # Retrieve Context, Run and Fab for the run_id
+            serverapp_ctxt = state.get_serverapp_context(run_id)
+            run = state.get_run(run_id)
+            fab = None
+            if run and run.fab_hash:
+                if result := ffs.get(run.fab_hash):
+                    fab = Fab(run.fab_hash, result[0])
+            if run and fab and serverapp_ctxt:
+                # Update run status to STARTING
+                if state.update_run_status(run_id, RunStatus(Status.STARTING, "", "")):
+                    log(INFO, "Starting run %d", run_id)
+                    return PullServerAppInputsResponse(
+                        context=context_to_proto(serverapp_ctxt),
+                        run=run_to_proto(run),
+                        fab=fab_to_proto(fab),
+                    )
+
+        # Raise an exception if the Run or Fab is not found,
+        # or if the status cannot be updated to STARTING
+        raise RuntimeError(f"Failed to start run {run_id}")
+
+    def PushServerAppOutputs(
+        self, request: PushServerAppOutputsRequest, context: grpc.ServicerContext
+    ) -> PushServerAppOutputsResponse:
+        """Push ServerApp process outputs."""
+        log(DEBUG, "ServerAppIoServicer.PushServerAppOutputs")
+
+        # Init state
+        state = self.state_factory.state()
+
+        # Abort if the run is not running
+        abort_if(
+            request.run_id,
+            [Status.PENDING, Status.STARTING, Status.FINISHED],
+            state,
+            context,
+        )
+
+        state.set_serverapp_context(request.run_id, context_from_proto(request.context))
+        return PushServerAppOutputsResponse()
+
+    def UpdateRunStatus(
+        self, request: UpdateRunStatusRequest, context: grpc.ServicerContext
+    ) -> UpdateRunStatusResponse:
+        """Update the status of a run."""
+        log(DEBUG, "ServerAppIoServicer.UpdateRunStatus")
+
+        # Init state
+        state = self.state_factory.state()
+
+        # Abort if the run is finished
+        abort_if(request.run_id, [Status.FINISHED], state, context)
+
+        # Update the run status
+        state.update_run_status(
+            run_id=request.run_id, new_status=run_status_from_proto(request.run_status)
+        )
+        return UpdateRunStatusResponse()
+
+    def PushLogs(
+        self, request: PushLogsRequest, context: grpc.ServicerContext
+    ) -> PushLogsResponse:
+        """Push logs."""
+        log(DEBUG, "ServerAppIoServicer.PushLogs")
+        state = self.state_factory.state()
+
+        # Add logs to LinkState
+        merged_logs = "".join(request.logs)
+        state.add_serverapp_log(request.run_id, merged_logs)
+        return PushLogsResponse()
+
+    def GetRunStatus(
+        self, request: GetRunStatusRequest, context: grpc.ServicerContext
+    ) -> GetRunStatusResponse:
+        """Get the status of a run."""
+        log(DEBUG, "ServerAppIoServicer.GetRunStatus")
+        state = self.state_factory.state()
+
+        # Get run status from LinkState
+        run_statuses = state.get_run_status(set(request.run_ids))
+        run_status_dict = {
+            run_id: run_status_to_proto(run_status)
+            for run_id, run_status in run_statuses.items()
+        }
+        return GetRunStatusResponse(run_status_dict=run_status_dict)
+
+
+def _raise_if(validation_error: bool, request_name: str, detail: str) -> None:
+    """Raise a `ValueError` with a detailed message if a validation error occurs."""
+    if validation_error:
+        raise ValueError(f"Malformed {request_name}: {detail}")

flwr/server/superlink/ffs/__init__.py

@@ -0,0 +1,24 @@
+# Copyright 2024 Flower Labs GmbH. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Flower File Storage for large objects."""
+
+
+from .disk_ffs import DiskFfs as DiskFfs
+from .ffs import Ffs as Ffs
+
+__all__ = [
+    "DiskFfs",
+    "Ffs",
+]