Using python script instead of sh

.github/workflows/smoke_tests.yaml

@@ -27,6 +27,6 @@ jobs:
           pip install --upgrade pip
           pip install $(grep -v '^torchdata\|^torchtext\|^torcheval' requirements.txt)
       - name: Run Script
-        run: sh examples/utils/run_fl_local.sh # TODO create new script to check for logs, move it to the tests folder
+        run: python -m tests.smoke_tests.run_smoke_test
       - name: Clearing pip for space
         run: pip uninstall -y -r requirements.txt

README.md

@@ -148,7 +148,7 @@ We use the standard git development flow of branch and merge to main with PRs on
 
 The library dependencies and those for development are listed in the `pyproject.toml` and `requirements.txt` files. You may use whatever virtual environment management tool that you would like. These include conda, poetry, and virtualenv. Poetry is used to produce our releases, which are managed and automated by GitHub.
 
-The easiest way to create and activate a virtual environment is
+The easiest way to create and activate a virtual environment is by using the [virtualenv](https://pypi.org/project/virtualenv/) package:
 ```bash
 virtualenv "ENV_PATH"
 source "ENV_PATH/bin/activate"

tests/smoke_tests/config.yaml

@@ -0,0 +1,25 @@
+# Parameters that describe server
+n_server_rounds: 15 # The number of rounds to run FL
+
+# Following the adaptive proximal weight setup in Appendix C3.3 of the FedProx paper: https://arxiv.org/pdf/1812.06127.pdf
+# Server decides to increase or decrease the proximal weight based on the average training losses of clients that were sent during training.
+# The updated proximal weight is then sent to the clients for the next round of training.
+# Set the initial proximal weight to 0.0 for adaptive proximal weight setup.
+adaptive_proximal_weight: True # Whether to use adaptive proximal weight or not
+proximal_weight : 0.0 # Initial proximal weight
+proximal_weight_delta : 0.1 # The amount by which to increase or decrease the proximal weight, if adaptive_proximal_weight is True
+proximal_weight_patience : 5 # The number of rounds to wait before increasing or decreasing the proximal weight, if adaptive_proximal_weight is True
+
+# Parameters that describe clients
+n_clients: 3 # The number of clients in the FL experiment
+local_epochs: 1 # The number of epochs to complete for client
+batch_size: 128 # The batch size for client training
+
+reporting_config:
+  enabled: False
+  project_name: FL4Health # Name of the project under which everything should be logged
+  run_name: "FedProx Server" # Name of the run on the server-side, each client will also have it's own run name
+  group_name: "FedProx Experiment" # Group under which each of the FL run logging will be stored
+  entity: "your_entity_here" # WandB user name
+  notes: "Testing WB reporting"
+  tags: ["Test", "FedProx"]

tests/smoke_tests/run_smoke_test.py

@@ -0,0 +1,55 @@
+import logging
+import subprocess
+import time
+
+logging.basicConfig()
+logger = logging.getLogger()
+logger.setLevel(logging.INFO)
+
+
+def run_smoke_test(
+    n_clients_to_start: int = 4,
+    config_path: str = "tests/smoke_tests/config.yaml",
+    dataset_path: str = "examples/datasets/mnist_data/",
+) -> None:
+    # Start the server, divert the outputs to a server file
+    logger.info("Starting server")
+
+    ps = subprocess.Popen([
+        "nohup",
+        "python",
+        "-m",
+        "examples.fedprox_example.server",
+        "--config_path",
+        config_path,
+    ], stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
+    server_output = ps.stdout
+
+    # Sleep for 20 seconds to allow the server to come up.
+    # TODO fix this by capturing the output and parsing it
+    time.sleep(20)
+
+    # Start n number of clients and divert the outputs to their own files
+    client_outputs = []
+    for i in range(n_clients_to_start):
+        logger.info(f"Starting client {i}")
+        ps = subprocess.Popen([
+            "nohup",
+            "python",
+            "-m",
+            "examples.fedprox_example.client",
+            "--dataset_path",
+            dataset_path,
+        ], stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
+        client_outputs.append(ps.stdout)
+
+    # TODO make asserts
+
+    while True:
+        logger.info(f"Server output: {server_output.readline()}")
+        for i in range(len(client_outputs)):
+            logger.info(f"Client {i} output: {client_outputs[i].readline()}")
+
+
+if __name__ == "__main__":
+    run_smoke_test()