Dev/repair develop #270
Closed
Dev/repair develop #270
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| @@ -136,7 +139,8 @@ def inference(self, img: torch.tensor, interpolate: bool = False): | |||
| pad = int((W - H) / 2) | |||
| features = F.interpolate(features, new_size, mode="bilinear", align_corners=True) | |||
| features = F.pad(features, pad=[pad, pad, 0, 0]) | |||
|
|
|||
| # Optionally turn on image feature smoothing | |||
| # features = filter2d(features, self.mean_kernel, "replicate") | |||
There was a problem hiding this comment.
Did smoothening feature help in some cases?
Comment on lines
30
to
31
| std = 400 | ||
| mean = 550 |
There was a problem hiding this comment.
This we should set in the LossAnomalyParams. These parameters were found with a test dataset per environment. We should update this during online learning with a rolling window from the traversed features.
| @@ -26,13 +26,10 @@ segmentation_type: "random" | |||
| feature_type: "dino" | |||
| dino_patch_size: 16 # DINO only | |||
| confidence_std_factor: 4.0 | |||
| scale_traversability: True | |||
| scale_traversability: False # This parameter needs to be false when using the anomaly detection model | |||
There was a problem hiding this comment.
Should add logic to set to false automatically
| @@ -273,5 +360,16 @@ def load_model(self): | |||
| if __name__ == "__main__": | |||
| node_name = "wvn_feature_extractor_node" | |||
| rospy.init_node(node_name) | |||
|
|
|||
| if True: | |||
|
Working now with slic components: |
|
Closed
14 tasks
…visual_navigation into dev/repair_develop
Changes from deployment on robot
…isual_navigation into dev/repair_develop
Closed
…isual_navigation into dev/repair_develop
|
Working version with anomaly detection, running average loss computation and online learning
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Keeping track of changes:
Improved the logging of the feature_extraction node, which makes it easy to keep track if the node is up and running.
The pipeline is alive:
Slic features also works nicely:
