Skip to content

ONNX export does not work with Segment Everything #66

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kdubovikov wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −8
Open

ONNX export does not work with Segment Everything #66

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions export_to_onnx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,16 +38,16 @@ def export_onnx_esam(model, output):
onnx_model = onnx_models.OnnxEfficientSam(model=model)
dynamic_axes = {
"batched_images": {0: "batch", 2: "height", 3: "width"},
"batched_point_coords": {2: "num_points"},
"batched_point_labels": {2: "num_points"},
"batched_point_coords": {1: "num_points"},
"batched_point_labels": {1: "num_points"},
}
dummy_inputs = {
"batched_images": torch.randn(1, 3, 1080, 1920, dtype=torch.float),
"batched_point_coords": torch.randint(
low=0, high=1080, size=(1, 1, 5, 2), dtype=torch.float
low=0, high=1080, size=(1, 5, 1, 2), dtype=torch.float
),
"batched_point_labels": torch.randint(
low=0, high=4, size=(1, 1, 5), dtype=torch.float
low=0, high=4, size=(1, 5, 1), dtype=torch.float
),
}
output_names = ["output_masks", "iou_predictions"]
Expand Down Expand Up @@ -82,16 +82,16 @@ def export_onnx_esam_decoder(model, output):
onnx_model = onnx_models.OnnxEfficientSamDecoder(model=model)
dynamic_axes = {
"image_embeddings": {0: "batch"},
"batched_point_coords": {2: "num_points"},
"batched_point_labels": {2: "num_points"},
"batched_point_coords": {1: "num_points"},
"batched_point_labels": {1: "num_points"},
}
dummy_inputs = {
"image_embeddings": torch.randn(1, 256, 64, 64, dtype=torch.float),
"batched_point_coords": torch.randint(
low=0, high=1080, size=(1, 1, 5, 2), dtype=torch.float
low=0, high=1080, size=(1, 5, 1, 2), dtype=torch.float
),
"batched_point_labels": torch.randint(
low=0, high=4, size=(1, 1, 5), dtype=torch.float
low=0, high=4, size=(1, 5, 1), dtype=torch.float
),
"orig_im_size": torch.tensor([1080, 1920], dtype=torch.long),
}
Expand Down