Add basic parallel client example

clients/python/basic-parallel-client.py

@@ -0,0 +1,32 @@
+import argparse
+import umbridge
+import random
+import time
+from concurrent.futures import ThreadPoolExecutor
+
+
+# Read URL from command line argument
+parser = argparse.ArgumentParser(description='Parallel evaluations via UM-Bridge.')
+parser.add_argument('url', metavar='url', type=str,
+                    help='the URL at which the model is running, for example http://localhost:4242')
+args = parser.parse_args()
+print(f"Connecting to URL {args.url}")
+
+# Connect to model server
+model = umbridge.HTTPModel(args.url, "forward")
+
+def evaluate_model(number):
+    # Define parameter for evaluation
+    parameters = [[number for _ in range(model.get_input_sizes()[0])]]
+    # Evaluate model
+    model_output = model(parameters)
+    print(model_output)
+    return model_output
+
+# Number of evaluations to perform
+num_evaluations = 100
+
+# Run evaluations in parallel
+# Use max_workers to define the maximum number of threads
+with ThreadPoolExecutor(max_workers = 6) as executor:
+    results = list(executor.map(evaluate_model, range(1, num_evaluations + 1)))