Resampling » History » Version 1
Marcus Pearce, 2013-07-15 03:29 PM
| 1 | 1 | Marcus Pearce | h1. Resampling |
|---|---|---|---|
| 2 | 1 | Marcus Pearce | |
| 3 | 1 | Marcus Pearce | These are some low-level functions in the <code>resampling</code> package, which are called by the <code>idyom:idyom</code> function. You shouldn't need to use them unless you are developing the code. |
| 4 | 1 | Marcus Pearce | |
| 5 | 1 | Marcus Pearce | h2. <code>resampling:idyom-resample</code> |
| 6 | 1 | Marcus Pearce | |
| 7 | 1 | Marcus Pearce | <code>idyom:idyom</code> uses <code>resampling:idyom-resample</code> to compute the note-by-note IC values, and can be used to obtain these a list. The function takes a subset of the top-level arguments (see above): |
| 8 | 1 | Marcus Pearce | |
| 9 | 1 | Marcus Pearce | * Required: dataset-id, target-viewpoints, source-viewpoints (no viewpoint selection) |
| 10 | 1 | Marcus Pearce | * Model: models, ltmo, stmo |
| 11 | 1 | Marcus Pearce | * Training: pretraining-ids, k, resampling-indices |
| 12 | 1 | Marcus Pearce | |
| 13 | 1 | Marcus Pearce | h2. <code>resampling:output-information-content</code> |
| 14 | 1 | Marcus Pearce | |
| 15 | 1 | Marcus Pearce | Takes the output of <code>resampling:idyom-resample</code> and returns the average information content. It takes the following arguments: |
| 16 | 1 | Marcus Pearce | |
| 17 | 1 | Marcus Pearce | * predictions: the output of <code>resampling:idyom-resample</code> |
| 18 | 1 | Marcus Pearce | * detail: an integer which determines how the information content is averaged (these are returned as multiple values): |
| 19 | 1 | Marcus Pearce | ** 1: averaged over the entire dataset |
| 20 | 1 | Marcus Pearce | ** 2: and also averaged over each composition |
| 21 | 1 | Marcus Pearce | ** 3: and also for each event in each composition |
| 22 | 1 | Marcus Pearce | |
| 23 | 1 | Marcus Pearce | h2. <code>resampling:format-information-content</code> |
| 24 | 1 | Marcus Pearce | |
| 25 | 1 | Marcus Pearce | <code>resampling:format-information-content</code> takes the output of <code>resampling:idyom-resample</code> and writes it to file. It takes the following arguments: |
| 26 | 1 | Marcus Pearce | |
| 27 | 1 | Marcus Pearce | * predictions: the output of <code>resampling:idyom-resample</code> |
| 28 | 1 | Marcus Pearce | * file: a string denoting a file |
| 29 | 1 | Marcus Pearce | * dataset-id: an integer reflecting the dataset-id |
| 30 | 1 | Marcus Pearce | * detail: an integer which determines how the information content is averaged (these are returned as multiple values): |
| 31 | 1 | Marcus Pearce | ** 1: averaged over the entire dataset |
| 32 | 1 | Marcus Pearce | ** 2: and also averaged over each composition |
| 33 | 1 | Marcus Pearce | ** 3: and also for each event in each composition |