Upload 2 files

app.py

@@ -70,29 +70,29 @@ class_names = data_dict['names']  # e.g., ['Platelets', 'RBC', 'WBC'] if those a
 
 ##################################
 # Hardcoded metrics from your provided values:
-overall_precision = 0.870496260646755
-overall_recall = 0.8460399765524981
-overall_map50 = 0.9160845656283895
-overall_map = 0.6064155939296477
-overall_map75 = 0.6557004942673867
+overall_precision = 0.8998657967724281
+overall_recall = 0.9152413015416975
+overall_map50 = 0.9482967626275897
+overall_map = 0.6529025986330599
+overall_map75 = 0.7199225312247104
 
 # Per-Class Metrics (index as per data.yaml order)
 # Here we assume the class order matches the indices: 
 # class_names[0], class_names[1], class_names[2], etc.
-class0_precision = 0.7648872215366018
-class0_recall = 0.9452054794520548
-class0_map50 = 0.9106238743284377
-class0_map = 0.44560784653430324
-
-class1_precision = 0.8560449257059868
-class1_recall = 0.6329144502054396
-class1_map50 = 0.8441600369816818
-class1_map = 0.5859616928719056
-
-class2_precision = 0.9905566346976764
-class2_recall = 0.96
-class2_map50 = 0.9934697855750487
-class2_map = 0.7876772423827341
+class0_precision = 0.8820047185253768
+class0_recall = 0.958904109589041
+class0_map50 = 0.961433378998409
+class0_map = 0.5023530432704303
+
+class1_precision = 0.8196728808767741
+class1_recall = 0.7868197950360514
+class1_map50 = 0.8884569088843599
+class1_map = 0.6302822447945686
+
+class2_precision = 0.9979197909151334
+class2_recall = 1.0
+class2_map50 = 0.995
+class2_map = 0.8260725078341811
 
 # Construct the metrics DataFrame
 metrics_data = [

best.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4196097b0bfc16018a43ca683bd993f0a51cf352ac28edf76d17a4e49f4fe437
-size 5464787
+oid sha256:3e5903d2178c13636e443cdcee1af62f54d33e8de317ec2ce198d07bc4dfd98c
+size 5471059