-
Notifications
You must be signed in to change notification settings - Fork 5
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #119 from JamesOwers/amt
Amt
- Loading branch information
Showing
10 changed files
with
1,146 additions
and
20 deletions.
There are no files selected for viewing
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
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
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,149 @@ | ||
| """A degrader object can be used to easily degrade data points on the fly | ||
| according to some given parameters.""" | ||
| import json | ||
| import numpy as np | ||
| import warnings | ||
|
|
||
| import mdtk.degradations as degs | ||
|
|
||
| class Degrader(): | ||
| """A Degrade object can be used to easily degrade musical excerpts | ||
| on the fly.""" | ||
|
|
||
| def __init__(self, seed=None, degradations=list(degs.DEGRADATIONS.keys()), | ||
| degradation_dist=np.ones(len(degs.DEGRADATIONS)), | ||
| clean_prop=1 / (len(degs.DEGRADATIONS) + 1), config=None): | ||
| """ | ||
| Create a new degrader with the given parameters. | ||
| Parameters | ||
| ---------- | ||
| seed : int | ||
| A random seed for numpy. | ||
| degradations : list(string) | ||
| A list of the names of the degradations to use (and in what order | ||
| to label them). | ||
| degradation_dist : list(float) | ||
| A list of the probability of each degradation given in | ||
| degradations. This list will be normalized to sum to 1. | ||
| clean_prop : float | ||
| The proportion of degrade calls that should return clean excerpts. | ||
| config : string | ||
| The path of a json config file (created by measure_errors.py). | ||
| If given, degradations, degradation_dist, and clean_prop will | ||
| all be overwritten by the values in the json file. | ||
| """ | ||
| if seed is not None: | ||
| np.random.seed(seed) | ||
|
|
||
| # Load config | ||
| if config is not None: | ||
| with open(config, 'r') as file: | ||
| config = json.load(file) | ||
|
|
||
| if 'degradation_dist' in config: | ||
| degradation_dist = np.array(config['degradation_dist']) | ||
| degradations = list(degs.DEGRADATIONS.keys()) | ||
| if 'clean_prop' in config: | ||
| clean_prop = config['clean_prop'] | ||
|
|
||
| # Check arg validity | ||
| assert len(degradation_dist) == len(degradations), ( | ||
| "Given degradation_dist is not the same length as degradations:" | ||
| f"\nlen({degradation_dist}) != len({degradations})" | ||
| ) | ||
| assert min(degradation_dist) >= 0, ("degradation_dist values must " | ||
| "not be negative.") | ||
| assert sum(degradation_dist) > 0, ("Some degradation_dist value " | ||
| "must be positive.") | ||
| assert 0 <= clean_prop <= 1, ("clean_prop must be between 0 and 1 " | ||
| "(inclusive).") | ||
|
|
||
| self.degradations = degradations | ||
| self.degradation_dist = degradation_dist | ||
| self.clean_prop = clean_prop | ||
| self.failed = np.zeros(len(degradations)) | ||
|
|
||
|
|
||
| def degrade(self, note_df): | ||
| """ | ||
| Degrade the given note_df. | ||
| Parameters | ||
| ---------- | ||
| note_df : pd.DataFrame | ||
| A note_df to degrade. | ||
| Returns | ||
| ------- | ||
| degraded_df : pd.DataFrame | ||
| A degraded version of the given note_df. If self.clean_prop > 0, | ||
| this can be a copy of the given note_df. | ||
| deg_label : int | ||
| The label of the degradation that was performed. 0 means none, | ||
| and larger numbers mean the degradation | ||
| "self.degradations[deg_label+1]" was performed. | ||
| """ | ||
| if self.clean_prop > 0 and np.random.rand() <= self.clean_prop: | ||
| return note_df.copy(), 0 | ||
|
|
||
| degraded_df = None | ||
| this_deg_dist = self.degradation_dist.copy() | ||
| this_failed = self.failed.copy() | ||
|
|
||
| # First, sample from failed degradations | ||
| while np.any(this_failed > 0): | ||
| # Select a degradation proportional to how many have failed | ||
| deg_index = np.random.choice( | ||
| len(self.degradations), | ||
| p=this_failed / np.sum(this_failed) | ||
| ) | ||
| deg_fun = degs.DEGRADATIONS[self.degradations[deg_index]] | ||
|
|
||
| # Try to degrade | ||
| with warnings.catch_warnings(): | ||
| warnings.simplefilter("ignore") | ||
| degraded_df = deg_fun(note_df) | ||
|
|
||
| # Check for success! | ||
| if degraded_df is not None: | ||
| self.failed[deg_index] -= 1 | ||
| return degraded_df, deg_index + 1 | ||
|
|
||
| # Degradation failed -- 0 out this deg and continue | ||
| this_failed[deg_index] = 0 | ||
|
|
||
| # No degradations have remaining failures. Draw from standard dist | ||
| while np.any(this_deg_dist > 0): | ||
| # Select a degradation proportional to the distribution | ||
| deg_index = np.random.choice( | ||
| len(self.degradations), | ||
| p=this_deg_dist / np.sum(this_deg_dist) | ||
| ) | ||
| # This deg would have already failed in the above loop. | ||
| # But we want to sample it and count it as another failure. | ||
| if self.failed[deg_index] > 0: | ||
| self.failed[deg_index] += 1 | ||
| continue | ||
| deg_fun = degs.DEGRADATIONS[self.degradations[deg_index]] | ||
|
|
||
| # Try to degrade | ||
| with warnings.catch_warnings(): | ||
| warnings.simplefilter("ignore") | ||
| degraded_df = deg_fun(note_df) | ||
|
|
||
| # Check for success! | ||
| if degraded_df is not None: | ||
| return degraded_df, deg_index + 1 | ||
|
|
||
| # Degradation failed -- add 1 to failure and continue | ||
| self.failed[deg_index] += 1 | ||
|
|
||
| # Here, all degradations (with dist > 0) failed | ||
| return note_df.copy(), 0 | ||
|
|
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
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
Oops, something went wrong.