Update script.py

script.py

@@ -1,59 +1,72 @@
 import pandas as pd
 import numpy as np
-from PIL import Image
 import onnxruntime as ort
 import os
 from tqdm import tqdm
-
+import timm
+import torchvision.transforms as T
+from PIL import Image
+import torch
 
 def is_gpu_available():
     """Check if the python package `onnxruntime-gpu` is installed."""
-    return ort.get_device() == "GPU"
+    return torch.cuda.is_available()
 
 
-class ONNXWorker:
+class PytorchWorker:
     """Run inference using ONNX runtime."""
 
-    def __init__(self, onnx_path: str):
-        print("Setting up ONNX runtime session.")
-        self.use_gpu = is_gpu_available()
-        if self.use_gpu:
-            providers = ["CUDAExecutionProvider", "CPUExecutionProvider"]
-        else:
-            providers = ["CPUExecutionProvider"]
+    def __init__(self, model_path: str, model_name: str, number_of_categories: int = 1604):
+
+        def _load_model(model_name, model_path):
+
+            print("Setting up Pytorch Model")
+            self.device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+            print(f"Using devide: {self.device}")
+
+            model = timm.create_model(model_name, num_classes=number_of_categories, pretrained=False)
+
+            # if not torch.cuda.is_available():
+            #     model_ckpt = torch.load(model_path, map_location=torch.device("cpu"))
+            # else:
+            #     model_ckpt = torch.load(model_path)
+
+            model_ckpt = torch.load(model_path, map_location=self.device)
+            model.load_state_dict(model_ckpt)
+
+            return model.to(self.device).eval()
 
-        print(f"Using {providers}")
-        self.ort_session = ort.InferenceSession(onnx_path, providers=providers)
+        self.model = _load_model(model_name, model_path)
 
-    def _resize_image(self, image: np.ndarray) -> np.ndarray:
+        self.transforms = T.Compose([T.Resize((299, 299)),
+                                     T.ToTensor(),
+                                     T.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])
 
-        new_size = (384, 384) 
-        return np.array(Image.fromarray(image).resize(new_size))
-    
 
     def predict_image(self, image: np.ndarray) -> list():
+        """Run inference using ONNX runtime.
+        :param image: Input image as numpy array.
+        :return: A list with logits and confidences.
+        """
 
-        """Run inference using ONNX runtime."""
-        resized_image = self._resize_image(image)
-        logits = self.ort_session.run(None, {"input": resized_image})
+        logits = self.model(self.transforms(image).unsqueeze(0).to(self.device))
 
         return logits.tolist()
 
 
-def make_submission(test_metadata, model_path, output_csv_path="./submission.csv", images_root_path="/tmp/data/private_testset"):
+def make_submission(test_metadata, model_path, model_name, output_csv_path="./submission.csv", images_root_path="/tmp/data/private_testset"):
     """Make submission with given """
 
-    model = ONNXWorker(model_path)
+    model = PytorchWorker(model_path, model_name)
 
     predictions = []
 
     for _, row in tqdm(test_metadata.iterrows(), total=len(test_metadata)):
-        image_path = os.path.join(images_root_path, row.filename)
+        image_path = os.path.join(images_root_path, row.image_path)
 
         test_image = Image.open(image_path).convert("RGB")
-        test_image_resized = np.asarray(test_image.resize((384, 384)))
 
-        logits = model.predict_image(test_image_resized)
+        logits = model.predict_image(test_image)
 
         predictions.append(np.argmax(logits))
 
@@ -70,12 +83,14 @@ if __name__ == "__main__":
     with zipfile.ZipFile("/tmp/data/private_testset.zip", 'r') as zip_ref:
         zip_ref.extractall("/tmp/data")
 
-    ONNX_MODEL_PATH = "./MetaFG_meta_2.onnx"
+    MODEL_PATH = "pytorch_model.bin"
+    MODEL_NAME = "tf_efficientnet_b1.ap_in1k"
 
     metadata_file_path = "./SnakeCLEF2024-TestMetadata.csv"
     test_metadata = pd.read_csv(metadata_file_path)
 
     make_submission(
         test_metadata=test_metadata,
-        model_path=ONNX_MODEL_PATH,
-    )
+        model_path=MODEL_PATH,
+        model_name=MODEL_NAME
+    )