Replaced deprecated data types and fixed some typos in README.md #16
Conversation
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
Some nonexhaustive tarballs and various zip extensions.
- included updated environment and setup instructions - shifted data collection part into detail/summary blocks
Author
|
Hi @setarehc, I've found a few more things I thought might help make your project more awesome:
Hope you have time to review and maybe merge some of these :) |
- initially "text"/"action" was set early, so args.dataset was checked twice - we now set "action" by default with option to switch to "text" per dataset
- moved working notes into its own header section - renamed sections for training and inference
- updated "custom" in documentation to "myrig" - added hml_vec and dataset name in synthesis part - introduced new scripts for the custom workflow - build_vectors: compute rot_6d representation from custom rig - annotate_texts: create annotated texts from raw descriptions - cal_mean_variance: compute mean and variance for custom rig
- aligned the inference part - added preliminary motion_process details
- building the beginning of the inference part - added details for the training of custom models
- will need to update njoints manually - added custom to encode_text()
- dropped some KIT references
- restructured text - added pointers to the new class needed in dataset.py - added the start of the output shaping section - moved some exploratory notes into instruction section - removed edit_args dump since TrainingArgs blocks explain better
- replaced unused HumanML3D class with modified CustomRig template class - updated t2m minimum animation length to 5 frames - updated n_joints from 22 to 27 in conversion functions - removed block containing just comments - added explicit type and arg messages for exceptions - added CustomRig imports in get_data.py and utils/model_util.py
- added steps in general sequence at top of README.md - data_loaders/custom_utils.py for custom rig matrix initialization - utils/editing_util.py for new mask computation - configs/card.py for overriding options - custom_utils.py: new matrix dimensions - editing_util.py: joint_to_full_mask_custom() - card.py: new card dataclass - corresponding training steps in train/train_condmdi.py
- commented out unused evaluation import - added catch for ZeroDivisionError in data loading
- changed paths for data/text loading - calculated dim_pose based on humanml_opt.txt
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.
Based on some warnings, I removed np.float/bool etc and replaced them with the non-numpy versions. Also I fixed some things in README which I thought were typos, hope it helps.