support CUDA Place

paddle_fl/core/trainer/fl_trainer.py

@@ -57,11 +57,11 @@ class FLTrainer(object):
         self._current_ep = None
         self.cur_step = 0
 
-    def start(self):
+    def start(self, place):
         #current_ep = "to be added"
         self.agent = FLWorkerAgent(self._scheduler_ep, self._current_ep)
         self.agent.connect_scheduler()
-        self.exe = fluid.Executor(fluid.CPUPlace())
+        self.exe = fluid.Executor(place)
         self.exe.run(self._startup_program)
 
     def run(self, feed, fetch):
@@ -103,11 +103,11 @@ class FedAvgTrainer(FLTrainer):
         super(FedAvgTrainer, self).__init__()
         pass
 
-    def start(self):
+    def start(self, place):
         #current_ep = "to be added"
         self.agent = FLWorkerAgent(self._scheduler_ep, self._current_ep)
         self.agent.connect_scheduler()
-        self.exe = fluid.Executor(fluid.CPUPlace())
+        self.exe = fluid.Executor(place)
         self.exe.run(self._startup_program)
 
     def set_trainer_job(self, job):
@@ -185,10 +185,10 @@ class SecAggTrainer(FLTrainer):
     def step_id(self, s):
         self._step_id = s
 
-    def start(self):
+    def start(self, place):
         self.agent = FLWorkerAgent(self._scheduler_ep, self._current_ep)
         self.agent.connect_scheduler()
-        self.exe = fluid.Executor(fluid.CPUPlace())
+        self.exe = fluid.Executor(place)
         self.exe.run(self._startup_program)
         self.cur_step = 0
 

paddle_fl/examples/ctr_demo/fl_trainer.py

@@ -11,7 +11,7 @@
 # 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.
-
+import paddle.fluid as fluid
 from paddle_fl.core.trainer.fl_trainer import FLTrainerFactory
 from paddle_fl.core.master.fl_job import FLRunTimeJob
 import numpy as np
@@ -42,7 +42,8 @@ job.load_trainer_job(job_path, trainer_id)
 job._scheduler_ep = "127.0.0.1:9091"  # Inform the scheduler IP to trainer
 trainer = FLTrainerFactory().create_fl_trainer(job)
 trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 print(trainer._scheduler_ep, trainer._current_ep)
 output_folder = "fl_model"
 epoch_id = 0

paddle_fl/examples/dpsgd_demo/fl_trainer.py

@@ -35,7 +35,8 @@ job.load_trainer_job(job_path, trainer_id)
 job._scheduler_ep = "127.0.0.1:9091"  # Inform scheduler IP address to trainer
 trainer = FLTrainerFactory().create_fl_trainer(job)
 trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 
 test_program = trainer._main_program.clone(for_test=True)
 

paddle_fl/examples/femnist_demo/fl_trainer.py

@@ -37,7 +37,8 @@ job._scheduler_ep = "127.0.0.1:9091"  # Inform the scheduler IP to trainer
 print(job._target_names)
 trainer = FLTrainerFactory().create_fl_trainer(job)
 trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 print(trainer._step)
 test_program = trainer._main_program.clone(for_test=True)
 

paddle_fl/examples/gru4rec_demo/fl_trainer.py

@@ -36,7 +36,8 @@ job.load_trainer_job(job_path, trainer_id)
 job._scheduler_ep = "127.0.0.1:9091"  # Inform the scheduler IP to trainer
 trainer = FLTrainerFactory().create_fl_trainer(job)
 trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 
 r = Gru4rec_Reader()
 train_reader = r.reader(train_file_dir, place, batch_size=125)

paddle_fl/examples/k8s_deployment/trainer0/fl_trainer.py

@@ -11,7 +11,7 @@
 # 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.
-
+import paddle.fluid as fluid
 from paddle_fl.core.trainer.fl_trainer import FLTrainerFactory
 from paddle_fl.core.master.fl_job import FLRunTimeJob
 import numpy as np
@@ -47,7 +47,8 @@ trainer = FLTrainerFactory().create_fl_trainer(job)
 #trainer._current_ep = "127.0.0.1:{}".format(9000+trainer_id)
 trainer._current_ep = os.environ['TRAINER0_SERVICE_HOST'] + ":" + os.environ[
     'TRAINER0_SERVICE_PORT_TRAINER0']
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 print(trainer._scheduler_ep, trainer._current_ep)
 output_folder = "fl_model"
 epoch_id = 0

paddle_fl/examples/k8s_deployment/trainer1/fl_trainer.py

@@ -11,7 +11,7 @@
 # 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.
-
+import paddle.fluid as fluid
 from paddle_fl.core.trainer.fl_trainer import FLTrainerFactory
 from paddle_fl.core.master.fl_job import FLRunTimeJob
 import numpy as np
@@ -47,7 +47,8 @@ trainer = FLTrainerFactory().create_fl_trainer(job)
 #trainer._current_ep = "127.0.0.1:{}".format(9000+trainer_id)
 trainer._current_ep = os.environ['TRAINER1_SERVICE_HOST'] + ":" + os.environ[
     'TRAINER1_SERVICE_PORT_TRAINER1']
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 print(trainer._scheduler_ep, trainer._current_ep)
 output_folder = "fl_model"
 epoch_id = 0

paddle_fl/examples/secagg_demo/fl_trainer.py

@@ -53,7 +53,8 @@ trainer.trainer_id = trainer_id
 trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
 trainer.trainer_num = trainer_num
 trainer.key_dir = "./keys/"
-trainer.start()
+place = fluid.CPUPlace()
+trainer.start(place)
 
 output_folder = "fl_model"
 epoch_id = 0

paddle_fl/examples/submitter_demo/train_program.py

@@ -99,7 +99,8 @@ else:
     job._scheduler_ep = scheduler_conf["ENDPOINT"]
     trainer = FLTrainerFactory().create_fl_trainer(job)
     trainer._current_ep = endpoint
-    trainer.start()
+    place = fluid.CPUPlace()
+    trainer.start(place)
     print(trainer._scheduler_ep, trainer._current_ep)
     output_folder = "fl_model"
     epoch_id = 0