pivtools 0.1.3__cp311-cp311-win_amd64.whl

pivtools_gui/pivtools.py

@@ -0,0 +1,58 @@
+import tracemalloc
+
+import dask
+import dask.config
+from dask.distributed import Client, LocalCluster
+
+from .config import Config
+from .image_handling.load_images import load_images
+from .image_handling.readers import list_supported_formats
+from .post_processing.POD.pod_decompose import pod_decompose, pod_rebuild
+from pivtools_cli.preprocessing.preprocess import preprocess_images
+from .vector_statistics.ensemble_statistics import ensemble_statistics
+from .vector_statistics.instantaneous_statistics import instantaneous_statistics
+
+tracemalloc.start()
+
+
+if __name__ == "__main__":
+    # Print supported image formats
+    print(f"Supported image formats: {', '.join(list_supported_formats())}")
+
+    # Start a local Dask cluster and client for distributed computation and dashboard
+    cluster = LocalCluster()
+    client = Client(cluster)
+    print(f"Dask dashboard available at: {client.dashboard_link}")
+
+    # Print cluster resources
+    print(f"Number of workers: {len(client.scheduler_info()['workers'])}")
+    print("Threads per worker:")
+    for addr, worker in client.scheduler_info()["workers"].items():
+        print(f"  {addr}: {worker['nthreads']} threads")
+
+    dask.config.set(scheduler="threads")
+
+    config = Config()
+    base_paths = config.base_paths
+    sources = config.source_paths
+    camera_numbers = config.camera_numbers
+
+    for idx, base_path in enumerate(base_paths):
+        source = sources[idx]
+        for cam_num in camera_numbers:
+            print(f"Processing source: {source}, camera: {cam_num}")
+            images = load_images(cam_num, config, source=source)
+            processed_images = preprocess_images(images, config)
+            processed_images.compute()
+            # perform PIV
+            instantaneous_statistics(cam_num, config, base=base_path)
+            ensemble_statistics(cam_num, config, base=base_path)
+            pod_decompose(cam_num, config, base=base_path, k_modes=10)
+            # Rebuild calibrated fields at prescribed energy (if configured)
+            pod_rebuild(cam_num, config, base=base_path)
+
+    current, peak = tracemalloc.get_traced_memory()
+    print(f"Current memory usage: {current / 10**6:.2f} MB")
+    print(f"Peak memory usage: {peak / 10**6:.2f} MB")
+
+    tracemalloc.stop()