Merge commit '5e58474c'

README.md

@@ -5,7 +5,7 @@ infos on current collaborations in
 https://docs.google.com/document/d/179dD1d6lmWhQ9e2E1AUoJyLZ1d_5c-aMNwAPkbmIcBU
 
 ### metadata.py
-Stores a dictionary of datasets and characteristics (SR, NFFT, and path to access soundfiles)
+Stores a dictionary of datasets and characteristics (SR, NFFT, path to access soundfiles, and downsampling factor for ultra/infra-sonic signals)
 Convenient to iterate over the whole dataset
 ```
 from metadata import species
@@ -26,33 +26,40 @@ Runs all baseline algorithms over the dataset.
 - [x] pyin (librosa implem)
 - [x] crepe (torchcrepe implem)
 - [x] crepe finetuned (torchcrepe implem)
-- [ ] crepe finetuned over all species BUT the one it is evaluated on
+- [x] crepe finetuned over all species except the target
+- [x] crepe finetuned only on the target species
 - [x] crepe (original tensorflow implem https://arxiv.org/abs/1802.06182)
 - [x] basic pitch (https://arxiv.org/abs/2203.09893)
 - [x] pesto (https://arxiv.org/abs/2309.02265)
+- [~] pesto finetuned over the target species
 
 This scripts stores predictions along with resampled annotations in `{basename}_preds.csv` files
 
+### print_annot.py
+For each vocalisation, prints a spectrogram and overlaid annotations and predictions as .png file stored in the `pred_pngs` folder.
+
 ### eval_all.py
 Evaluates each algorithms over the dataset using `{basename}_preds.csv` files, with a threshold of 50 cents for accuracies.
 For each algorithms and species, this outputs ROC optimal thresholds, Recall, False alarm, Pitch accuracy, and Chroma accuracy.
 /!\ These metrics are mesured per vocalisation before being averaged.
 Scores are stored in `scores/{specie}_scores.csv` files
 
-### get_noisy_labels.py
-Detects potential misannotations.
-It measures
-- the SNR as the median of the annotated f0's energy relative to the median energy for its time bin.
-- the presence of a sub-harmonic as the SNR (see above) of half the annotated f0
-- the presence of a 2/3 sub-harmonic as the SNR (see above) of 2/3 the annotated f0
-These values are then thresholded and "noisy" vocalisation spectrograms are copied into the `noisy_pngs` folder to browse and check results. One can then check if the spectrogram exists in `noisy_pngs` to discard noisy labels.
+### compute_salience_SHR.py
+Evaluates metrics for each annotated temporal bin:
+- the presence of a sub-harmonic following [this paper](https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=cb8f47c23c74932152456a6f7a464fd3a2321259)
+- the saliency of the annotation as the ratio of the energy of the f0 (one tone around the annotation) and its surrounding (one octave around the annotation)
+These values are stored in the SHR and salience columns of the {basename}_preds.csv files
+
+### get_noisy_pngs.py
+Thresholds saliency and SHR values vocalisation wise and copies those considered "noisy" into the `noisy_pngs` folder to browse and check results. One can then check if the spectrogram exists in `noisy_pngs` to discard noisy labels.
 
 ### train_crepe.py
 Fine tunes the crepe model using the whole dataset.
 - [x] Loads 1024 sample windows and their corresponding f0 to be stored in a large `train_set.pkl` file (skip if data hasn't changed).
 - [x] Applies gradient descent using the BCE following the crepe paper (this task is treated as a binary classification for each spectral bin).
-- [x] The fine tuned model is stored in `model_all.pth`
-- [ ] Train on all species but one given as argument
+- [x] The fine tuned model is stored in `crepe_ft/model_all.pth`
+- [x] Train on one target species given as argument (weights are stored in `crepe_ft/model_only_{specie}.pth)
+- [x] Train on all species except the target given as argument (weights are stored in `crepe_ft/model_omit_{specie}.pth)