Upload folder using huggingface_hub

README.md

@@ -1,4 +1,14 @@
-# My Cool Submission 2025
+# Empty solution example for the S23DR competition
 
-This repo contains a submission to the [S23DR Challenge](https://huggingface.co/spaces/usm3d/S23DR) (part of the [USM3D](https://usm3d.github.io/) workshop at CVPR2025). It was prepared by [jacklangerman](https://huggingface.co/jacklangerman).
+This repo provides a minimalistic example of a valid, but empty submission to S23DR competition.
+We recommend you take a look at [this example](https://huggingface.co/usm3d/handcrafted_baseline_submission),
+which implements some primitive algorithms and provides useful I/O and visualization functions.
+
+This example seeks to simply provide minimal code which succeeds at reading the dataset and producing a solution (in this case two vertices at the origin and edge of zero length connecting them).
+
+`script.py` - is the main file which is run by the competition space. It should produce `submission.parquet` as the result of the run. Please see the additional comments in the `script.py` file. 
+
+---
+license: apache-2.0
+---
 

script.py

@@ -52,8 +52,11 @@ if __name__ == "__main__":
 
     if data_path_test_server.exists():
         data_path = data_path_test_server
+        TEST_ENV = True
     else: 
         data_path = data_path_local
+        TEST_ENV = False
+
     
     print(data_path)
 
@@ -65,22 +68,23 @@ if __name__ == "__main__":
     print(data_files)
 
 
-    try:
+    if TEST_ENV:
         dataset = load_dataset(
-            "arrow", 
+            "webdataset", 
             data_files=data_files,
             trust_remote_code=True,
             # streaming=True
         )
-        print('load with arrow')
-    except:
+        print('load with webdataset')
+    else:
         dataset = load_dataset(
-            "webdataset", 
+            "arrow", 
             data_files=data_files,
             trust_remote_code=True,
             # streaming=True
         )
-        print('load with webdataset')
+        print('load with arrow')
+    
 
     print(dataset, flush=True)
     # dataset = load_dataset('webdataset', data_files={)