onnx

configs/inference.yaml

@@ -8,7 +8,7 @@ checkpoint_path: "./checkpoints/checkpoint.pth"
 
 input:
   # For video processing
-  video_path: "face.mp4"
+  video_path: "M2Ohb0FAaJU_1.mp4"
   frame_skip: 0
 
   # For single frame processing

onnxconv.py

@@ -0,0 +1,108 @@
+import torch
+import torch.onnx
+from IMF.model import IMFModel
+
+
+
+def trace_handler(module, input, output):
+    print(f"\nModule: {module.__class__.__name__}")
+    for idx, inp in enumerate(input):
+        print_tensor_info(inp, f"  Input[{idx}]")
+    if isinstance(output, torch.Tensor):
+        print_tensor_info(output, "  Output")
+    else:
+        for idx, out in enumerate(output):
+            print_tensor_info(out, f"  Output[{idx}]")
+
+
+def print_tensor_info(tensor, name, indent=0):
+    indent_str = ' ' * indent
+    print(f"{indent_str}{name}:")
+    if isinstance(tensor, torch.Tensor):
+        print(f"{indent_str}  Shape: {tensor.shape}")
+        print(f"{indent_str}  Dtype: {tensor.dtype}")
+        print(f"{indent_str}  Device: {tensor.device}")
+        print(f"{indent_str}  Requires grad: {tensor.requires_grad}")
+    elif isinstance(tensor, (list, tuple)):
+        print(f"{indent_str}  Type: {type(tensor).__name__}, Length: {len(tensor)}")
+        for idx, item in enumerate(tensor):
+            print_tensor_info(item, f"{name}[{idx}]", indent=indent + 2)
+    else:
+        print(f"{indent_str}  Type: {type(tensor).__name__}")
+
+def print_model_structure(model):
+    print("Model Structure:")
+    for name, module in model.named_modules():
+        print(f"{name}: {module.__class__.__name__}")
+        if hasattr(module, 'weight'):
+            print(f"  Weight shape: {module.weight.shape}")
+        if hasattr(module, 'bias') and module.bias is not None:
+            print(f"  Bias shape: {module.bias.shape}")
+
+def export_to_onnx(model, x_current, x_reference, file_name):
+    try:
+        print("Model structure before tracing:")
+        print_model_structure(model)
+
+        print("\nInput tensor information:")
+        print_tensor_info(x_current, "x_current")
+        print_tensor_info(x_reference, "x_reference")
+
+        hooks = []
+        for name, module in model.named_modules():
+            hooks.append(module.register_forward_hook(trace_handler))
+
+        # Use torch.jit.trace to create a traced version of the model
+        print("\nTracing model...")
+        traced_model = torch.jit.trace(model, (x_current, x_reference))
+        print("Model traced successfully")
+
+        for hook in hooks:
+            hook.remove()
+
+        print("\nModel structure after tracing:")
+        print_model_structure(traced_model)
+
+        print("\nExporting to ONNX...")
+        torch.onnx.export(
+            model,
+            (x_current, x_reference),
+            "imf_model.onnx",
+            export_params=True,
+            opset_version=11,
+            do_constant_folding=True,
+            input_names=['x_current', 'x_reference'],
+            output_names=['output'],
+            dynamic_axes={
+                'x_current': {0: 'batch_size'},
+                'x_reference': {0: 'batch_size'},
+                'output': {0: 'batch_size'}
+            },
+            verbose=True
+        )
+
+
+        print(f"Model exported successfully to {file_name}")
+    except Exception as e:
+        print(f"Error during ONNX export: {str(e)}")
+        import traceback
+        traceback.print_exc()
+
+# Load your model
+model = IMFModel()
+model.eval()
+
+# Load the checkpoint
+checkpoint = torch.load("./checkpoints/checkpoint.pth", map_location=lambda storage, loc: storage)
+state_dict = checkpoint['model_state_dict']
+
+# # Adjust the weights in the state_dict
+# for key in state_dict.keys():
+#     if 'csonv.weight' in key and state_dict[key].dim() == 5:
+#         state_dict[key] = state_dict[key].squeeze(0)
+# Create dummy input tensors
+x_current = torch.randn(1, 3, 256, 256)
+x_reference = torch.randn(1, 3, 256, 256)
+
+# Export the model
+export_to_onnx(model, x_current, x_reference, "imf_model.onnx")