comparison scripts/map_and_average.py @ 62:4425a4918102 branch-tests

fixed indices for feature components

     '''
     trainset = load_data_from_pickle(INPUT_FILES[0])
     valset = load_data_from_pickle(INPUT_FILES[1])
     testset = load_data_from_pickle(INPUT_FILES[2])
     return trainset, valset, testset
+
+
+def limit_to_n_seconds(dataset, n_sec=30.0, win_sec=8.0):
+    X, Y, Yaudio = dataset
+    uniq_audio, uniq_counts = np.unique(Yaudio, return_counts=True)
+    frame_sr = 2.0
+    max_n_frames = np.int(np.floor((n_sec - win_sec) * frame_sr))
+    X_new, Y_new, Yaudio_new = [], [], []
+    for audio in uniq_audio:
+        idx = np.where(Yaudio==audio)[0]
+        if len(idx) > max_n_frames:
+            idx = idx[:max_n_frames]
+        X_new.append(X[idx, :])
+        Y_new.append(Y[idx])
+        Yaudio_new.append(Yaudio[idx])
+    return [np.concatenate(X_new), np.concatenate(Y_new), np.concatenate(Yaudio_new)]
 
 
 def get_feat_inds(n_dim=840):
     '''assume frame with 840 features and return indices for each feature
     '''