fix typos add singapore pipeline

cr_analysis.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-#SBATCH --output=cr_garden2.out
+#SBATCH --output=./logs/cr_garden2.out
 
 source ~/miniforge3/bin/activate
 
@@ -8,12 +8,11 @@ CONDA_ENV_PATH="~/moth_detector_env/"
 
 # Activate the environment
 conda activate "${CONDA_ENV_PATH}"
-#conda install --yes --file requirements.txt
 
 # Print the Costa Rica deployments avaialble on the object store
 # python print_deployments.py --subset_countries 'Costa Rica'
 
-# Run the Python script on JASMIN
+# Run the Inference script 
 python s3_download_with_inference.py \
   --country "Costa Rica" \
   --deployment "Garden - 3F1C4908"

s3_download_with_inference.py

@@ -261,7 +261,6 @@ def download_and_inference(
     print("\033[93m\033[1m" + "Pipeline parameters" + "\033[0m\033[0m")
     print(f"\033[93m - Scratch and crops storage: {data_storage_path}\033[0m")
 
-    crops_interval = args.crops_interval
     if args.keep_crops:
         crops_interval = args.crops_interval
         print(f"\033[93m - Keeping crops every {crops_interval}mins\033[0m")
@@ -272,7 +271,7 @@ def download_and_inference(
     download_and_inference(
         args.country,
         args.deployment,
-        crops_interval,
+        int(crops_interval),
         args.rerun_existing,
         data_storage_path,
         args.perform_inference,

singapore_analysis.sh

@@ -0,0 +1,23 @@
+#!/bin/bash
+#SBATCH --output=./logs/singapore.out
+#SBATCH --time=24:00:00
+
+source ~/miniforge3/bin/activate
+
+# Define the path to your existing Conda environment (modify as appropriate)
+CONDA_ENV_PATH="~/moth_detector_env/"
+
+# Activate the environment
+conda activate "${CONDA_ENV_PATH}"
+
+# Print the Costa Rica deployments avaialble on the object store
+# python print_deployments.py --subset_countries 'Costa Rica'
+
+# Run the Inference script 
+python s3_download_with_inference.py \
+  --country "Singapore" \
+  --deployment "All" \
+  --crops_interval 10 \
+  --keep_crops \
+  --data_storage_path ./data/singapore
+

utils/custom_models.py

@@ -85,38 +85,31 @@ def load_models(device, localisation_model_path, binary_model_path, order_model_
             in_features, num_classes
         )
     )
-    checkpoint = torch.load(weights_path, map_location=device)
+    checkpoint = torch.load(weights_path, map_location=device, weights_only=True)
     state_dict = checkpoint.get("model_state_dict") or checkpoint
     model_loc.load_state_dict(state_dict)
     model_loc = model_loc.to(device)
     model_loc.eval()
 
     # Load the binary model
     weights_path = binary_model_path
-    # labels_path = "/bask/homes/f/fspo1218/amber/data/mila_models/05-moth-nonmoth_category_map.json"
-    num_classes = 2  # moth, non-moth
     classification_model = timm.create_model(
         "tf_efficientnetv2_b3", num_classes=num_classes, weights=None
     )
     classification_model = classification_model.to(device)
-    checkpoint = torch.load(weights_path, map_location=device)
+    checkpoint = torch.load(weights_path, map_location=device, weights_only=True)
     state_dict = checkpoint.get("model_state_dict") or checkpoint
     classification_model.load_state_dict(state_dict)
     classification_model.eval()
 
     # Load the order model
     savedWeights = order_model_path
     thresholdFile = order_threshold_path
-    # img_size = 128
     order_data_thresholds = pd.read_csv(thresholdFile)
     order_labels = order_data_thresholds["ClassName"].to_list()
-    # thresholds = data_thresholds["Threshold"].to_list()
-    # means = data_thresholds["Mean"].to_list()
-    # stds = data_thresholds["Std"].to_list()
-    # img_depth = 3
     num_classes = len(order_labels)
     model_order = ResNet50_order(num_classes=num_classes)
-    model_order.load_state_dict(torch.load(savedWeights, map_location=device))
+    model_order.load_state_dict(torch.load(savedWeights, map_location=device, weights_only=True))
     model_order = model_order.to(device)
     model_order.eval()
 
@@ -130,7 +123,7 @@ def load_models(device, localisation_model_path, binary_model_path, order_model_
     num_classes = len(species_category_map)
     species_model = Resnet50_species(num_classes=num_classes)
     species_model = species_model.to(device)
-    checkpoint = torch.load(weights, map_location=device)
+    checkpoint = torch.load(weights, map_location=device, weights_only=True)
     # The model state dict is nested in some checkpoints, and not in others
     state_dict = checkpoint.get("model_state_dict") or checkpoint
     species_model.load_state_dict(state_dict)