updates

examples/simulation-pytorch/sim.ipynb

@@ -205,6 +205,7 @@
    "source": [
     "from datasets import Dataset\n",
     "from flwr_datasets import FederatedDataset\n",
+    "from datasets.utils.logging import disable_progress_bar\n",
     "\n",
     "# Let's set a simulation involving a total of 100 clients\n",
     "NUM_CLIENTS = 100\n",
@@ -568,13 +569,18 @@
     "# client needs exclusive access to these many resources in order to run\n",
     "client_resources = {\"num_cpus\": 1, \"num_gpus\": 0.0}\n",
     "\n",
+    "# Let's disable tqdm progress bar in the main thread (used by the server)\n",
+    "disable_progress_bar()\n",
     "\n",
     "history = fl.simulation.start_simulation(\n",
     "    client_fn=client_fn_callback,  # a callback to construct a client\n",
     "    num_clients=NUM_CLIENTS,  # total number of clients in the experiment\n",
     "    config=fl.server.ServerConfig(num_rounds=10),  # let's run for 10 rounds\n",
     "    strategy=strategy,  # the strategy that will orchestrate the whole FL pipeline\n",
     "    client_resources=client_resources,\n",
+    "    actor_kwargs={\n",
+    "        \"on_actor_init_fn\": disable_progress_bar  # disable tqdm on each actor/process spawning virtual clients\n",
+    "    },\n",
     ")"
    ]
   },

examples/simulation-pytorch/sim.py

@@ -10,10 +10,10 @@
 from flwr.common.typing import Scalar
 
 from datasets import Dataset
+from datasets.utils.logging import disable_progress_bar
 from flwr_datasets import FederatedDataset
 
-from utils import Net, train, test, get_mnist_transforms
-
+from utils import Net, train, test, mnist_transforms
 
 parser = argparse.ArgumentParser(description="Flower Simulation with PyTorch")
 
@@ -101,11 +101,12 @@ def client_fn(cid: str) -> fl.client.Client:
 
         # Now we apply the transform to each batch.
         trainset = trainset.map(
-            lambda img: {"img": get_mnist_transforms()(img)}, input_columns="image"
+            lambda img: {"image": mnist_transforms(img)},
+            input_columns="image",
         ).with_format("torch")
 
         valset = valset.map(
-            lambda img: {"img": get_mnist_transforms()(img)}, input_columns="image"
+            lambda img: {"image": mnist_transforms(img)}, input_columns="image"
         ).with_format("torch")
 
         # Create and return client
@@ -160,9 +161,12 @@ def evaluate(
 
         # Apply transform to dataset
         testset = centralized_testset.map(
-            lambda img: {"img": get_mnist_transforms()(img)}, input_columns="image"
+            lambda img: {"image": mnist_transforms(img)}, input_columns="image"
         ).with_format("torch")
 
+        # Disable tqdm for dataset preprocessing
+        disable_progress_bar()
+
         testloader = DataLoader(testset, batch_size=50)
         loss, accuracy = test(model, testloader, device=device)
 
@@ -206,6 +210,9 @@ def main():
         client_resources=client_resources,
         config=fl.server.ServerConfig(num_rounds=args.num_rounds),
         strategy=strategy,
+        actor_kwargs={
+            "on_actor_init_fn": disable_progress_bar  # disable tqdm on each actor/process spawning virtual clients
+        },
     )
 
 

examples/simulation-pytorch/utils.py

@@ -5,6 +5,10 @@
 from torchvision.transforms import ToTensor, Normalize, Compose
 
 
+# transformation to convert images to tensors and apply normalization
+mnist_transforms = Compose([ToTensor(), Normalize((0.1307,), (0.3081,))])
+
+
 # Model (simple CNN adapted from 'PyTorch: A 60 Minute Blitz')
 class Net(nn.Module):
     def __init__(self, num_classes: int = 10) -> None:
@@ -33,7 +37,7 @@ def train(net, trainloader, optim, epochs, device: str):
     net.train()
     for _ in range(epochs):
         for batch in trainloader:
-            images, labels = batch["img"].to(device), batch["label"].to(device)
+            images, labels = batch["image"].to(device), batch["label"].to(device)
             optim.zero_grad()
             loss = criterion(net(images), labels)
             loss.backward()
@@ -48,19 +52,10 @@ def test(net, testloader, device: str):
     net.eval()
     with torch.no_grad():
         for data in testloader:
-            images, labels = data["img"].to(device), data["label"].to(device)
+            images, labels = data["image"].to(device), data["label"].to(device)
             outputs = net(images)
             loss += criterion(outputs, labels).item()
             _, predicted = torch.max(outputs.data, 1)
             correct += (predicted == labels).sum().item()
     accuracy = correct / len(testloader.dataset)
     return loss, accuracy
-
-
-def get_mnist_transforms():
-    """Get transformation for MNIST dataset."""
-
-    # transformation to convert images to tensors and apply normalization
-    tr = Compose([ToTensor(), Normalize((0.1307,), (0.3081,))])
-
-    return tr

examples/simulation-tensorflow/sim.ipynb

@@ -116,7 +116,7 @@
    "outputs": [],
    "source": [
     "class FlowerClient(fl.client.NumPyClient):\n",
-    "    def __init__(self,  trainset, valset) -> None:\n",
+    "    def __init__(self, trainset, valset) -> None:\n",
     "        # Create model\n",
     "        self.model = get_model()\n",
     "        self.x_train, self.y_train = trainset[\"image\"], trainset[\"label\"]\n",
@@ -189,7 +189,6 @@
     "\n",
     "        trainset = client_dataset_splits[\"train\"].with_format(\"tf\")\n",
     "        valset = client_dataset_splits[\"test\"].with_format(\"tf\")\n",
-    "    \n",
     "\n",
     "        # Create and return client\n",
     "        return FlowerClient(trainset, valset)\n",
@@ -219,7 +218,9 @@
     "    ):\n",
     "        model = get_model()  # Construct the model\n",
     "        model.set_weights(parameters)  # Update model with the latest parameters\n",
-    "        loss, accuracy = model.evaluate(testset[\"image\"], testset[\"label\"], verbose=VERBOSE)\n",
+    "        loss, accuracy = model.evaluate(\n",
+    "            testset[\"image\"], testset[\"label\"], verbose=VERBOSE\n",
+    "        )\n",
     "        return loss, {\"accuracy\": accuracy}\n",
     "\n",
     "    return evaluate"
@@ -330,18 +331,6 @@
    "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.9.16"
   }
  },
  "nbformat": 4,

examples/simulation-tensorflow/sim.py

@@ -92,7 +92,6 @@ def client_fn(cid: str) -> fl.client.Client:
 
         trainset = client_dataset_splits["train"].with_format("tf")
         valset = client_dataset_splits["test"].with_format("tf")
-
 
         # Create and return client
         return FlowerClient(trainset, valset)
@@ -137,7 +136,9 @@ def evaluate(
     ):
         model = get_model()  # Construct the model
         model.set_weights(parameters)  # Update model with the latest parameters
-        loss, accuracy = model.evaluate(testset["image"], testset["label"], verbose=VERBOSE)
+        loss, accuracy = model.evaluate(
+            testset["image"], testset["label"], verbose=VERBOSE
+        )
         return loss, {"accuracy": accuracy}
 
     return evaluate