Open
Conversation
Add link to example data.
…nterpolate to isotropic in load_data. Linear matching reward function.
…=False in draw_line_segment.
…layer. Loop through reward hyperparameters in training for comparison.
Track with branching
…and resolved bugs. - Updated the `sample_from_output` function to use the correct device for the covariance matrix in the MultivariateNormal distribution. - Removed CPU detachment in the actor output during the update process. - Enhanced memory sampling in the training loop to include a transformation option. - Modified the `inference` function to include a `save` parameter, allowing control over saving inference results.
…and resolved bugs. - Updated the `sample_from_output` function to use the correct device for the covariance matrix in the MultivariateNormal distribution. - Removed CPU detachment in the actor output during the update process. - Enhanced memory sampling in the training loop to include a transformation option. - Modified the `inference` function to include a `save` parameter, allowing control over saving inference results.
- In `branch_classifier.py`, replaced image loading method from `tf.imread` to `torch.load`, and commented out unused code for clarity. Added a TODO for online data sampling from SWC and image files. Updated the `train` function to include a new structure for handling training data. - In `sac.py`, added handling for NaN values in the output of the `sample_from_output` function. Modified the `train` function to include a `training` flag for environment steps. Enhanced the `inference` function to support synchronization of processed images and improved path handling for saving results.
- In `branch_classifier.py`, replaced image loading method from `tf.imread` to `torch.load`, and commented out unused code for clarity. Added a TODO for online data sampling from SWC and image files. Updated the `train` function to include a new structure for handling training data. - In `sac.py`, added handling for NaN values in the output of the `sample_from_output` function. Modified the `train` function to include a `training` flag for environment steps. Enhanced the `inference` function to support synchronization of processed images and improved path handling for saving results.
… converage to estimated return.
…eurons with width.
- Modify environment and training functions to combine multiple seed points into a single episode. - Create python scripts to process neuron data (crop and scale images and swc points) and setup RL training input folder. - Create tree.py module to remove soma from swc files and simplify neuron trees. - Add random width, random noise, and random brightness options to simulated neuron drawing. - Simplify collect random points function for branch data collection and fix issue with out-of-bounds random points. - Create intensity inhomogeneity correction function in data_utils.py - Fix segment drawing function so consecutive segments connect without a gap and use interpolation for start and end points. - Handle loading swc files that contain all floating point values. - Modify parse_swc to parse swc lists with disconnected sections
- Implement loading of target networks and log_alpha during model initialization. - Add optimizer state loading for Q-networks and actor. - Update setup script to handle optional image directory and save neuron density maps. - Modify neuron drawing functions to support mask-based drawing. - Refactor show_state function for improved visualization and added live update capability during inference. - Remove deprecated show_state script. - Improve error handling and user prompts during inference.
- Updated the `show_state` function in `tracking_interface.py` to set the `vmax` and `vmin` parameters for better image contrast when displaying the RGB MIP. - Modified the alpha loss calculation in `sac.py` to directly use `log_alpha` instead of its exponential, improving numerical stability. - Adjusted the condition for determining `true_neuron` in the `inference` function to simplify the logic and enhance clarity.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
No description provided.