Merge branch 'dev/v0.7.0' of https://github.com/FedML-AI/FedML into dev/v0.7.0

Author: alexliang

android/fedmlsdk/src/main/java/ai/fedml/edge/FedEdgeImpl.java

@@ -63,7 +63,7 @@ public void handleMessage(Message msg) {
  } else if (MSG_TRAIN_ACCURACY == msg.what) {
  Bundle bundle = msg.getData();
  if (onTrainProgressListener != null && bundle != null) {
- onTrainProgressListener.onEpochLoss(msg.arg1, msg.arg2,
+ onTrainProgressListener.onEpochAccuracy(msg.arg1, msg.arg2,
  bundle.getFloat(TRAIN_ACCURACY, 0));
  }
  }

python/examples/cross_silo/mqtt_s3_fedavg_defense_mnist_lr_example/config/foolsgold/fedml_config.yaml

@@ -11,11 +11,11 @@ environment_args:
 data_args:
  dataset: "cifar10"
  data_cache_dir: ~/fedml_data
- partition_method: "hetero"
+ partition_method: "homo"
  partition_alpha: 0.5
 
 model_args:
- model: "lr"
+ model: "resnet56"
  model_file_cache_folder: "./model_file_cache" # will be filled by the server automatically
  global_model_file_path: "./model_file_cache/global_model.pt"
 
@@ -24,8 +24,8 @@ train_args:
  # for CLI running, this can be None; in MLOps deployment, `client_id_list` will be replaced with real-time selected devices
  client_id_list:
  # for FoolsGold Defense, if use_memory is true, then client_num_in_total should be equal to client_number_per_round
- client_num_in_total: 1000
- client_num_per_round: 2
+ client_num_in_total: 8
+ client_num_per_round: 8
  comm_round: 10
  epochs: 1
  batch_size: 10
@@ -37,16 +37,13 @@ validation_args:
  frequency_of_the_test: 1
 
 device_args:
- worker_num: 2
- using_gpu: false
- gpu_mapping_file: config/gpu_mapping.yaml
- gpu_mapping_key: mapping_config3_11
+ worker_num: 8
+ using_gpu: true
+ gpu_mapping_file: config/foolsgold/gpu_mapping.yaml
+ gpu_mapping_key: mapping_default
 
 comm_args:
- backend: "MQTT_S3"
- mqtt_config_path:
- s3_config_path:
- grpc_ipconfig_path: ./config/grpc_ipconfig.csv
+ backend: "MPI"
 
 tracking_args:
  # the default log path is at ~/fedml-client/fedml/logs/ and ~/fedml-server/fedml/logs/
@@ -64,4 +61,3 @@ attack_args:
 defense_args:
  enable_defense: true
  defense_type: foolsgold
- use_memory: true

python/examples/cross_silo/mqtt_s3_fedavg_defense_mnist_lr_example/config/foolsgold/gpu_mapping.yaml

@@ -0,0 +1,3 @@
+# this is used for 4 clients and 1 server training within a single machine which has 8 GPUs, but you hope to skip the GPU device ID.
+mapping_default:
+ host1: [3, 2, 2, 2] # assume we only have 4 GPUs

python/examples/cross_silo/mqtt_s3_fedavg_defense_mnist_lr_example/run_client.sh

@@ -1,4 +1,4 @@
 #!/usr/bin/env bash
 RANK=$1
 RUN_ID=$2
-python3 torch_client.py --cf config/krum/fedml_config.yaml --rank $RANK --role client --run_id $RUN_ID
+python3 torch_client.py --cf config/foolsgold/fedml_config.yaml --rank $RANK --role client --run_id $RUN_ID

python/examples/cross_silo/mqtt_s3_fedavg_defense_mnist_lr_example/run_mpi.sh

@@ -0,0 +1,12 @@
+#!/usr/bin/env bash
+
+WORKER_NUM=$1
+
+PROCESS_NUM=`expr $WORKER_NUM + 1`
+echo $PROCESS_NUM
+
+hostname > mpi_host_file
+
+mpirun -np $PROCESS_NUM \
+-hostfile mpi_host_file \
+python torch_mpi.py --cf config/foolsgold/fedml_config.yaml

python/examples/cross_silo/mqtt_s3_fedavg_defense_mnist_lr_example/torch_mpi.py

@@ -0,0 +1,31 @@
+import logging
+
+import fedml
+from fedml import FedMLRunner
+from fedml.model.cv.resnet import resnet56
+
+
+def create_model():
+ # please download the pre-trained weight file from
+ # https://github.com/chenyaofo/pytorch-cifar-models/releases/download/resnet/cifar10_resnet44-2a3cabcb.pt
+ pre_trained_model_path = "./config/resnet56_on_cifar10.pth"
+ model = resnet56(10, pretrained=True, path=pre_trained_model_path)
+ logging.info("load pretrained model successfully")
+ return model
+
+
+if __name__ == "__main__":
+ args = fedml.init()
+
+ # init device
+ device = fedml.device.get_device(args)
+
+ # load data
+ dataset, output_dim = fedml.data.load(args)
+
+ # load model
+ model = create_model()
+
+ # start training
+ fedml_runner = FedMLRunner(args, device, dataset, model)
+ fedml_runner.run()

python/fedml/core/security/defense/foolsgold_defense.py

@@ -15,6 +15,7 @@
 
 class FoolsGoldDefense(BaseDefenseMethod):
  def __init__(self, config):
+ super().__init__(config)
  self.config = config
  self.memory = None
 

python/fedml/core/security/defense/three_sigma_defense.py

@@ -162,9 +162,9 @@ def _get_importance_feature(self, raw_client_grad_list):
 
  # Get last key-value tuple
  (weight_name, importance_feature) = list(grads.items())[-2]
- print(importance_feature)
+ # print(importance_feature)
  feature_len = np.array(
- importance_feature.data.detach().numpy().shape
+ importance_feature.cpu().data.detach().numpy().shape
  ).prod()
  feature_vector = np.reshape(
  importance_feature.cpu().data.detach().numpy(), feature_len
@@ -195,14 +195,14 @@ def fools_gold_score(feature_vec_list):
  alpha[alpha <= 0.0] = 1e-15
 
  # Rescale so that max value is alpha
- print(np.max(alpha))
+  # print(np.max(alpha))
  alpha = alpha / np.max(alpha)
  alpha[(alpha == 1.0)] = 0.999999
 
  # Logit function
  alpha = np.log(alpha / (1 - alpha)) + 0.5
- alpha[(np.isinf(alpha) + alpha > 1)] = 1
- alpha[(alpha < 0)] = 0
+ # alpha[(np.isinf(alpha) + alpha > 1)] = 1
+ # alpha[(alpha < 0)] = 0
 
  print("alpha = {}".format(alpha))