add timm

Author: Liyulingyue

graph_net/test/timm_model_test.py

@@ -0,0 +1,55 @@
+# pip install timm
+
+import argparse
+import os
+import torch
+from torchvision import transforms
+import timm  # 导入 timm 库
+import graph_net
+
+os.environ["TIMMDL_DISABLE_RETRY"] = "1"  # 禁用重试
+
+
+def extract_visio_graph(model_name, model_path, dynamic_mode):
+    # Normalization parameters for ImageNet
+    normalize = transforms.Normalize(
+        mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]
+    )
+
+    # Create dummy input
+    batch_size = 1
+    height, width = 224, 224  # Standard ImageNet size
+    num_channels = 3
+    random_input = torch.rand(batch_size, num_channels, height, width)
+    normalized_input = normalize(random_input)
+
+    # Instantiate model using timm
+    model = timm.create_model(model_path, pretrained=False)  # 使用 timm 加载 resnet18
+    model.eval()
+
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    model.to(device)
+    normalized_input = normalized_input.to(device)
+
+    # Extract graph structure
+    model = graph_net.torch.extract(name=model_name, dynamic=dynamic_mode)(model)
+
+    print("Running inference...")
+    output = model(normalized_input)
+    print("Inference finished. Output shape:", output.shape)
+
+
+if __name__ == "__main__":
+    # get parameters from command line
+    workspace_default = os.environ.get("GRAPH_NET_EXTRACT_WORKSPACE", "workspace")
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--model_name", type=str, default="resnet18")
+    parser.add_argument("--model_path", type=str, default="resnet18")  # timm 模型名称
+    parser.add_argument("--workspace", type=str, default=workspace_default)
+    parser.add_argument("--dynamic", type=bool, default=True)
+    args = parser.parse_args()
+
+    os.environ["GRAPH_NET_EXTRACT_WORKSPACE"] = args.workspace
+
+    extract_visio_graph(args.model_name, args.model_path, args.dynamic)