WIP: Transfer learning implementation#10
Open
ftherrien wants to merge 20 commits into
Open
Conversation
…that were repeated (gas vars moved to gas area in last commit)
Co-authored-by: Copilot <copilot@github.com>
- add dataset-specific input/output column lists with validation - update gas/bicarb loaders to enforce column order and numeric cleanup - use dataset defaults in optimizer/run_experiment to avoid Voltage as input - make PhModel accept dynamic input size and Zero_eps_thickness index
…sistant between the liq and gas model. Now file name is found in YAML
…ther criteria in run_experiment
Streamline the data loading process by removing hardcoded column lists and replacing them with explicit column selection in loaders.py. This also includes cleaning up the GDEOptimizer and experiment pipeline by removing MAE (Mean Absolute Error) tracking and calculation, which was redundant with existing loss metrics. - Remove `default_input_labels` and `default_output_labels` from loaders.py - Hardcode specific input/output labels in `GDEOptimizer` for the gas dataset - Remove MAE calculation logic from `GDEOptimizer.get_predictor` - Remove MAE tracking from `run_active_learning_experiment` - Update `run_experiment.py` to use positional sys.argv for config loading - Add `data_file` to the default configuration schema
fix(gde_multi): prevent numerical instability and overflows Implement several safeguards to improve the numerical stability of the electrochemical model: - Clamp Butler-Volmer exponents using `_BV_MAX` to prevent infinite exponential values. - Replace `cosh`/`sinh` functions with exponential-based formulations to avoid overflow during large `M` calculations. - Add clamping to `c03` concentration and `current_density` to prevent division by zero or extreme values. - Use `torch.where` for Faraday efficiency calculations to safely handle near-zero current densities.
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
2 participants
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.
Implements transfer model from one model to another.
This should be rebased and merged after liquid_model is merged