Update nellie.py

nellie.py

@@ -120,11 +120,13 @@ def prepare_datasets(raw_tif_files, base_dir="generated_datasets"):
         
     return dirs
 
-def run_nellie_experiment(dataset_dir, combined_data_path, run_native_filter):
-    files = sorted(glob.glob(os.path.join(dataset_dir, "*.TIF")))
+def run_nellie_experiment(dataset_dir, combined_data_path, run_native_filter, target_filenames):
+    # EXPLICITLY use only the filenames we generated/sampled to prevent evaluating old data
+    files = [os.path.join(dataset_dir, fname) for fname in target_filenames]
     results, saved_masks = [], {}
     
     for filepath in tqdm(files, desc=f"Evaluating", leave=False):
+        if not os.path.exists(filepath): continue
         filename = os.path.basename(filepath)
         gt_mask_path = os.path.join(combined_data_path, filename.replace(".TIF", "_seg.npy"))
         if not os.path.exists(gt_mask_path): continue
@@ -220,20 +222,23 @@ def plot_single_experiment_visual(sample_filename, combined_data_path, data_src_
 # Main Execution Flow
 # =====================================================================
 def main():
-    n_samples = 10  # Default set to 10
+    n_samples = 10  # Evaluate only 10 samples
     
     logger.info("Downloading dataset...")
     dataset_dir = snapshot_download(repo_id="champ7/CEllDataW3", repo_type="dataset", allow_patterns="combined_data/*")
     combined_data_path = os.path.join(dataset_dir, "combined_data")
     
-    # Randomly sample N files to drastically speed up testing
+    # Extract complete list and sample N files
     raw_tif_files = sorted(glob.glob(os.path.join(combined_data_path, "*.TIF")))
     if n_samples and n_samples < len(raw_tif_files):
         random.seed(42) # Ensure the same random subset is used across runs
         raw_tif_files = random.sample(raw_tif_files, n_samples)
         logger.info(f"Subsampled dataset to {n_samples} random images.")
         
-    sample_filename = os.path.basename(raw_tif_files[0]) # File to plot for each step
+    # Isolate just the filenames to pass to the evaluation loop
+    target_filenames = [os.path.basename(f) for f in raw_tif_files]
+    
+    sample_filename = target_filenames[0] # File to plot for each step
     dirs = prepare_datasets(raw_tif_files)
     
     configs = [
@@ -252,10 +257,12 @@ def main():
     for config in configs:
         logger.info(f"Running strategy: {config['name']}")
         
+        # We now pass target_filenames to explicitly govern which files are read
         metrics, saved_masks = run_nellie_experiment(
             dataset_dir=dirs[config["data_src"]], 
             combined_data_path=combined_data_path, 
-            run_native_filter=config["run_filter"]
+            run_native_filter=config["run_filter"],
+            target_filenames=target_filenames
         )
         
         ious, precs, recs, f1s = [m[0] for m in metrics], [m[1] for m in metrics], [m[2] for m in metrics], [m[3] for m in metrics]
@@ -289,7 +296,7 @@ def main():
 
     # Print Final Summary Table
     print("\n" + "=" * 115)
-    print(" " * 35 + f"FINAL 7-WAY EVALUATION REPORT (N={len(raw_tif_files)} images)")
+    print(" " * 35 + f"FINAL 7-WAY EVALUATION REPORT (N={len(target_filenames)} images)")
     print("=" * 115)
     print(tabulate(aggregate_table, headers=headers, tablefmt="fancy_grid"))