Pipeline filenames - bugfix (#63)

* rename decollage -> flowcam, separate out filename parsing

* infer experiment name from directory, or supply, and passthru

* some debugging changes left over from last year

* back-incompatible ruff format pickiness, sigh

scripts/params.yaml

@@ -1,4 +1,4 @@
 cluster:
   n_clusters: 5
 
-collection: untagged-images-lana 
+collection: untagged-images-lana

scripts/run_luigi_pipeline.py

@@ -3,12 +3,52 @@
 from cyto_ml.pipeline.pipeline_decollage import FlowCamPipeline
 from cyto_ml.data.flowcam import parse_filename
 import logging
+<<<<<<< HEAD
+
+=======
+>>>>>>> main
 logging.basicConfig(level=logging.INFO)
 
 
 if __name__ == "__main__":
 
     parser = argparse.ArgumentParser(
+<<<<<<< HEAD
+        prog="Luigi FlowCam pipeline",
+        description="Triggers the process of decollaging images, embedding spatio-temporal metadata in headers, and uploading to s3 storage",
+    )
+
+    parser.add_argument(
+        "-o",
+        "--output_directory",
+        type=str,
+        default="./data/images_decollage",
+        help="directory for intermediate data",
+    )
+    parser.add_argument(
+        "-d",
+        "--directory",
+        type=str,
+        default="./tests/fixtures/MicrobialMethane_MESO_Tank10_54.0143_-2.7770_04052023_1",
+        help="Enable verbose mode",
+    )
+    parser.add_argument("-s", "--s3_bucket", type=str, default="untagged-images-lana")
+    parser.add_argument("-e", "--experiment_name", type=str)
+    args = parser.parse_args()
+
+    experiment = args.experiment_name
+    if not experiment or experiment == "":
+        try:
+            prefix, _, _, date, _ = parse_filename(args.directory)
+            experiment = "".join([prefix, date])
+        except ValueError as err:
+            logging.info(
+                "Could't figure out experiment name and date from {args.directory} - please call this with --experiment_name to set this"
+            )
+            logging.debug(err)
+            # TODO this properly
+            experiment = str(args.directory).split("/")[-2]
+=======
     prog='Luigi FlowCam pipeline',
     description='Triggers the process of decollaging images, embedding spatio-temporal metadata in headers, and uploading to s3 storage',
     )
@@ -28,6 +68,7 @@
             logging.info("Could't figure out experiment name and date from {args.directory} - please call this with --experiment_name to set this")
             logging.debug(err)
             exit
+>>>>>>> main
 
     luigi.build(
         [

src/cyto_ml/visualisation/app.py

@@ -168,7 +168,7 @@ def main() -> None:
     st.set_page_config(layout="wide", page_title="Plankton image embeddings")
 
     st.title("Image embeddings")
-    st.write(f"{len(image_ids(st.session_state['collection']))} images in {st.session_state["collection"]}")
+    st.write(f"{len(image_ids(st.session_state['collection']))} images in {st.session_state['collection']}")
     # the generated HTML is not lovely at all
 
     st.selectbox(