Skip to content
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.

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

init commit of trajectoryformer #7

Merged
merged 6 commits into from
Aug 22, 2023
Merged

init commit of trajectoryformer #7

merged 6 commits into from
Aug 22, 2023

Conversation

Cedarch
Copy link
Contributor

@Cedarch Cedarch commented Aug 18, 2023

No description provided.

Copy link
Collaborator

@poodarchu poodarchu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Make sure all .py files satisfy the codestyle check in CI.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do not include ad-hoc in library.
the key "pred" / "future" are specific terms in your impl, so keep these modifications in the experiment directory.

annos = ref_obj["objects"]
num_points_in_gt = np.array([ann["num_points"] for ann in annos])
gt_boxes = np.array([ann["box"] for ann in annos]).reshape(-1, 9)
difficulty = np.array([ann["detection_difficulty_level"] for ann in annos])

gt_ids = np.array([ann["name"] for ann in annos])
gt_names = np.array([TYPE_LIST[ann["label"]] for ann in annos])
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is gt_ids and gt_names the same?

"sweeps": [],
"frame_time": int(ref_obj['frame_name'].split("_")[-1]),
"context_name": ref_obj['scene_name']
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

make the key name the same as items in ref_obj?

gt_names = np.array([TYPE_LIST[ann["label"]] for ann in annos])
mask_not_zero = (num_points_in_gt > 0).reshape(-1)

# filter boxes without lidar points
annos_dict = {}
annos_dict["gt_boxes"] = gt_boxes[mask_not_zero, :].astype(np.float32)
annos_dict["gt_names"] = gt_names[mask_not_zero].astype(str)
annos_dict["gt_ids"] = gt_ids[mask_not_zero]
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

gt_ids & gt_names?

"gt_names": None,
"difficulty": None,
"name": None,
"num_points_in_gt": None
}
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

gt_boxes, gt_names, diffuculty, num_points_in_gt should be put in an "annotations" dict same as key-frame.
L178 has saved gt_boxes as annotations, why put them here again?

@Cedarch Cedarch requested a review from poodarchu August 18, 2023 16:39
Copy link
Collaborator

@poodarchu poodarchu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@poodarchu poodarchu merged commit 8f4be1f into V2AI:master Aug 22, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants