Enable the "--split_files" option in the gui

birdnet_analyzer/analyze/core.py

@@ -160,13 +160,10 @@ def analyze(
             output = audio_input
 
     if split_tables:
-        output: Path = (
-            audio_input_path.parent if audio_input_path.is_file() else audio_input_path
-        )
         _split_tables(
             df,
             audio_input_path,
-            output,
+            Path(output),
             fmin,
             fmax,
             predictions,
@@ -181,53 +178,53 @@ def analyze(
             sensitivity,
             species_list_file,
         )
+    else:
+        if "table" in rtypes:
+            save_as_rtable(
+                df,
+                fmin,
+                fmax,
+                predictions.model_fmin,
+                predictions.model_fmax,
+                audio_speed,
+                Path(output) / cfg.OUTPUT_RAVEN_FILENAME,
+            )
 
-    if "table" in rtypes:
-        save_as_rtable(
-            df,
-            fmin,
-            fmax,
-            predictions.model_fmin,
-            predictions.model_fmax,
-            audio_speed,
-            Path(output) / cfg.OUTPUT_RAVEN_FILENAME,
-        )
-
-    if "csv" in rtypes:
-        save_as_csv(
-            df,
-            Path(output) / cfg.OUTPUT_CSV_FILENAME,
-            additional_columns,
-            lat=lat,
-            lon=lon,
-            week=week,
-            overlap=overlap,
-            min_conf=min_conf,
-            sensitivity=sensitivity,
-            species_list_file=species_list_file,
-            model_path=predictions.model_path,
-        )
+        if "csv" in rtypes:
+            save_as_csv(
+                df,
+                Path(output) / cfg.OUTPUT_CSV_FILENAME,
+                additional_columns,
+                lat=lat,
+                lon=lon,
+                week=week,
+                overlap=overlap,
+                min_conf=min_conf,
+                sensitivity=sensitivity,
+                species_list_file=species_list_file,
+                model_path=predictions.model_path,
+            )
 
-    if "kaleidoscope" in rtypes:
-        save_as_kaleidoscope(df, Path(output) / cfg.OUTPUT_KALEIDOSCOPE_FILENAME)
+        if "kaleidoscope" in rtypes:
+            save_as_kaleidoscope(df, Path(output) / cfg.OUTPUT_KALEIDOSCOPE_FILENAME)
 
-    if "audacity" in rtypes:
-        save_as_audacity(df, Path(output) / cfg.OUTPUT_AUDACITY_FILENAME)
+        if "audacity" in rtypes:
+            save_as_audacity(df, Path(output) / cfg.OUTPUT_AUDACITY_FILENAME)
 
-    if "parquet" in rtypes:
-        save_as_parquet(
-            df,
-            Path(output) / cfg.OUTPUT_PARQUET_FILENAME,
-            additional_columns,
-            lat=lat,
-            lon=lon,
-            week=week,
-            overlap=overlap,
-            min_conf=min_conf,
-            sensitivity=sensitivity,
-            species_list_file=species_list_file,
-            model_path=predictions.model_path,
-        )
+        if "parquet" in rtypes:
+            save_as_parquet(
+                df,
+                Path(output) / cfg.OUTPUT_PARQUET_FILENAME,
+                additional_columns,
+                lat=lat,
+                lon=lon,
+                week=week,
+                overlap=overlap,
+                min_conf=min_conf,
+                sensitivity=sensitivity,
+                species_list_file=species_list_file,
+                model_path=predictions.model_path,
+            )
 
     if save_params:
         save_params_to_file(
@@ -355,6 +352,21 @@ def _split_tables(
                 df_file, output / (file_shorthand + ".BirdNET.results.txt")
             )
 
+        if "parquet" in rtypes:
+            save_as_parquet(
+                df_file,
+                output / (file_shorthand + ".BirdNET.results.parquet"),
+                additional_columns,
+                lat=lat,
+                lon=lon,
+                week=week,
+                overlap=overlap,
+                min_conf=min_conf,
+                sensitivity=sensitivity,
+                species_list_file=species_list_file,
+                model_path=predictions.model_path,
+            )
+
 
 def _merge_consecutive_segments(
     df: pd.DataFrame, merge_consecutive: int, hop_size: float = 3.0

birdnet_analyzer/gui/analysis.py

@@ -64,6 +64,7 @@ def run_analysis(
     n_producers,
     n_workers,
     progress: gr.Progress | None,
+    split_tables: bool = False,
 ):
     """Starts the analysis.
 
@@ -94,6 +95,7 @@ def run_analysis(
         n_workers: The number of worker threads to be used.
         input_dir: The input directory.
         progress: The gradio progress bar.
+        split_tables: Whether to split the output into separate tables per input file.
     """
     import birdnet_analyzer.gui.localization as loc
     from birdnet_analyzer.analyze import analyze
@@ -154,5 +156,6 @@ def run_analysis(
         save_params=save_params,
         n_producers=n_producers,
         n_workers=n_workers,
+        split_tables=split_tables,
         _return_only=bool(input_path),  # only for single file tab
     )

birdnet_analyzer/gui/multi_file.py

@@ -50,6 +50,7 @@ def run_batch_analysis(
     custom_classifier_file,
     output_type,
     additional_columns,
+    split_tables_checkbox,
     locale,
     batch_size,
     producers_number,
@@ -91,6 +92,7 @@ def run_batch_analysis(
         progress=progress,
         n_producers=producers_number,
         n_workers=workers_number,
+        split_tables=split_tables_checkbox,
     )
     skipped_files = [results.inputs[ui] for ui in results.unprocessable_inputs]
     header = (
@@ -219,6 +221,11 @@ def select_directory_wrapper():
                 label=loc.localize("multi-tab-additional-columns-checkbox-label"),
                 info=loc.localize("multi-tab-additional-columns-checkbox-info"),
             )
+            split_tables_checkbox = gr.Checkbox(
+                False,
+                label=loc.localize("multi-tab-split-table-checkbox-label"),
+                info=loc.localize("multi-tab-split-table-checkbox-info"),
+            )
 
         bs_number, producers_number, workers_number = gu.computing_settings()
         start_batch_analysis_btn = gr.Button(
@@ -247,6 +254,7 @@ def select_directory_wrapper():
             model_settings["selected_classifier_state"],
             output_type_radio,
             additional_columns_,
+            split_tables_checkbox,
             model_settings["locale_dropdown"],
             bs_number,
             producers_number,

birdnet_analyzer/lang/de.json

@@ -180,6 +180,8 @@
     "multi-tab-samples-dataframe-column-duration-header": "Länge",
     "multi-tab-samples-dataframe-column-subpath-header": "Unterpfad",
     "multi-tab-samples-dataframe-no-files-found": "Keine Dateien gefunden",
+    "multi-tab-split-table-checkbox-info": "Wenn aktiviert, wird für jede Eingabedatei eine separate Ergebnistabelle erstellt. Andernfalls werden alle Ergebnisse in einer einzigen Tabelle zusammengefasst.",
+    "multi-tab-split-table-checkbox-label": "Tabellen pro Eingabedatei aufteilen",
     "multi-tab-title": "Batch-Analyse",
     "progress-analyzing": "Analysiere",
     "progress-autotune": "Autotune läuft",
@@ -369,13 +371,13 @@
     "training-tab-upsampling-radio-option-smote": "SMOTE",
     "training-tab-upsampling-ratio-slider-info": "Das Mindestverhältnis für eine Minderheitenklasse im Vergleich zur Mehrheitsklasse nach dem Upsampling.",
     "training-tab-upsampling-ratio-slider-label": "Upsampling-Verhältnis",
-    "training-tab-warning-detached-classifier-append": "Der \"Anhängen\"-Modus wird für den abgekoppelten Klassifikator nicht angewendet.",
     "training-tab-use-focal-loss-checkbox-info": "Verwendet Focal Loss während des Trainings. Dies gibt schwierigen Beispielen mehr Gewicht und hilft, Klassenungleichgewichte zu bewältigen.",
     "training-tab-use-focal-loss-checkbox-label": "Focal Loss verwenden",
     "training-tab-use-labelsmoothing-checkbox-info": "Wendet Label Smoothing während des Trainings an. Dies kann die Generalisierung des Modells verbessern, indem es verhindert, dass das Modell zu selbstsicher wird.",
     "training-tab-use-labelsmoothing-checkbox-label": "Label Smoothing verwenden",
     "training-tab-use-mixup-checkbox-info": "Mixup ist eine Technik zur Datenvermehrung, die neue Daten durch Mischen zweier Beispiele und ihrer Label erzeugt.",
     "training-tab-use-mixup-checkbox-label": "Mixup verwenden",
+    "training-tab-warning-detached-classifier-append": "Der \"Anhängen\"-Modus wird für den abgekoppelten Klassifikator nicht angewendet.",
     "validation-max-confidence-lower-than-min-confidence": "Maximaler Konfidenz-Schwellwert muss größer sein als der minimale Schwellwert.",
     "validation-no-audio-directory-selected": "Kein Audioverzeichnis ausgewählt",
     "validation-no-cache-file-selected": "Bitte wählen Sie eine Cache-Datei aus.",
@@ -390,4 +392,4 @@
     "validation-no-valid-epoch-number": "Bitte geben Sie eine gültige Anzahl von Epochen an.",
     "validation-no-valid-frequency": "Bitte geben Sie eine gültige Frequenz an",
     "validation-no-valid-learning-rate": "Bitte geben Sie eine gültige Lernrate an."
-}
+}

birdnet_analyzer/lang/en.json

@@ -180,6 +180,8 @@
     "multi-tab-samples-dataframe-column-duration-header": "Length",
     "multi-tab-samples-dataframe-column-subpath-header": "Subpath",
     "multi-tab-samples-dataframe-no-files-found": "No files found",
+    "multi-tab-split-table-checkbox-info": "When enabled, a separate result table is created for each input file. Otherwise, all results are combined into a single table.",
+    "multi-tab-split-table-checkbox-label": "Split tables by input file",
     "multi-tab-title": "Batch analysis",
     "progress-analyzing": "Analyzing",
     "progress-autotune": "Autotune in progress",
@@ -369,13 +371,13 @@
     "training-tab-upsampling-radio-option-smote": "SMOTE",
     "training-tab-upsampling-ratio-slider-info": "The minimum ratio for a minority class compared to the majority class after upsampling.",
     "training-tab-upsampling-ratio-slider-label": "Upsampling ratio",
-    "training-tab-warning-detached-classifier-append": "Append mode will not apply to the detached classifier.",
     "training-tab-use-focal-loss-checkbox-info": "Uses focal loss during training. This gives more weight to hard examples and helps handle class imbalance.",
     "training-tab-use-focal-loss-checkbox-label": "Use focal loss",
     "training-tab-use-labelsmoothing-checkbox-info": "Applies label smoothing during training. This can help improve model generalization by preventing the model from becoming too confident.",
     "training-tab-use-labelsmoothing-checkbox-label": "Use label smoothing",
     "training-tab-use-mixup-checkbox-info": "Applies mixup augmentation during training. This can help improve model generalization by creating new training samples through linear interpolation.",
     "training-tab-use-mixup-checkbox-label": "Use mixup",
+    "training-tab-warning-detached-classifier-append": "Append mode will not apply to the detached classifier.",
     "validation-max-confidence-lower-than-min-confidence": "Maximum confidence must be greater than minimum confidence",
     "validation-no-audio-directory-selected": "No audio directory selected",
     "validation-no-cache-file-selected": "Please select a cache file.",

birdnet_analyzer/lang/fi.json

@@ -180,6 +180,8 @@
     "multi-tab-samples-dataframe-column-duration-header": "Pituus",
     "multi-tab-samples-dataframe-column-subpath-header": "Alihakemisto",
     "multi-tab-samples-dataframe-no-files-found": "Tiedostoja ei löytynyt",
+    "multi-tab-split-table-checkbox-info": "Kun tämä asetus on käytössä, jokaiselle syötetiedostolle luodaan oma tulostaulukko. Muussa tapauksessa kaikki tulokset yhdistetään yhteen taulukkoon.",
+    "multi-tab-split-table-checkbox-label": "Jaa taulukot syötetiedoston mukaan",
     "multi-tab-title": "Eräanalyysi",
     "progress-analyzing": "Analysoidaan",
     "progress-autotune": "Autoviritys käynnissä",
@@ -369,13 +371,13 @@
     "training-tab-upsampling-radio-option-smote": "SMOTE",
     "training-tab-upsampling-ratio-slider-info": "Vähemmistöluokan vähimmäissuhde enemmistöluokkaan verrattuna ylösnäytteistyksen jälkeen.",
     "training-tab-upsampling-ratio-slider-label": "Ylösnäytteistyssuhde",
-    "training-tab-warning-detached-classifier-append": "Liitä-tila ei koske irrotettua luokitinta.",
     "training-tab-use-focal-loss-checkbox-info": "Käyttää focal lossia koulutuksen aikana. Tämä antaa enemmän painoarvoa vaikeille esimerkeille ja auttaa käsittelemään luokkien epätasapainoa.",
     "training-tab-use-focal-loss-checkbox-label": "Käytä focal lossia",
     "training-tab-use-labelsmoothing-checkbox-info": "Soveltaa label smoothingia koulutuksen aikana. Tämä voi parantaa mallin yleistämistä estämällä mallia tulemasta liian itsevarmaksi.",
     "training-tab-use-labelsmoothing-checkbox-label": "Käytä label smoothingia",
     "training-tab-use-mixup-checkbox-info": "Sekoitus on datarikastusmenetelmä, joka luo uusia näytteitä yhdistämällä kaksi näytettä ja niiden leimaukset.",
     "training-tab-use-mixup-checkbox-label": "Käytä sekoitusta",
+    "training-tab-warning-detached-classifier-append": "Liitä-tila ei koske irrotettua luokitinta.",
     "validation-max-confidence-lower-than-min-confidence": "Enimmäisluotettavuuden on oltava suurempi kuin vähimmäisluotettavuus",
     "validation-no-audio-directory-selected": "Äänihakemistoa ei ole valittu",
     "validation-no-cache-file-selected": "Valitse välimuistitiedosto.",
@@ -390,4 +392,4 @@
     "validation-no-valid-epoch-number": "Anna kelvollinen aikakausien määrä.",
     "validation-no-valid-frequency": "Anna kelvollinen taajuus",
     "validation-no-valid-learning-rate": "Anna kelvollinen oppimistahti."
-}
+}

birdnet_analyzer/lang/fr.json

@@ -180,6 +180,8 @@
     "multi-tab-samples-dataframe-column-duration-header": "Longueur",
     "multi-tab-samples-dataframe-column-subpath-header": "Sous-chemin",
     "multi-tab-samples-dataframe-no-files-found": "Aucun fichiers trouvés",
+    "multi-tab-split-table-checkbox-info": "Lorsqu'elle est activée, une table de résultats distincte est créée pour chaque fichier d'entrée. Sinon, tous les résultats sont regroupés dans une seule table.",
+    "multi-tab-split-table-checkbox-label": "Séparer les tables par fichier d'entrée",
     "multi-tab-title": "Analyse par lots",
     "progress-analyzing": "Analyse en cours",
     "progress-autotune": "Autotune en progression",
@@ -369,13 +371,13 @@
     "training-tab-upsampling-radio-option-smote": "SMOTE",
     "training-tab-upsampling-ratio-slider-info": "Le ratio minimum pour une classe minoritaire par rapport à la classe majoritaire après le rééchantillonnage.",
     "training-tab-upsampling-ratio-slider-label": "Rapport de suréchantillonnage",
-    "training-tab-warning-detached-classifier-append": "Le mode d'ajout ne s'appliquera pas au classificateur détaché.",
     "training-tab-use-focal-loss-checkbox-info": "Utilise la perte focale pendant l'entraînement. Cela donne plus de poids aux exemples difficiles et aide à gérer le déséquilibre des classes.",
     "training-tab-use-focal-loss-checkbox-label": "Utiliser la perte focale",
     "training-tab-use-labelsmoothing-checkbox-info": "Applique le lissage des étiquettes pendant l'entraînement. Cela peut améliorer la généralisation du modèle en empêchant le modèle de devenir trop confiant.",
     "training-tab-use-labelsmoothing-checkbox-label": "Utiliser le lissage des étiquettes",
     "training-tab-use-mixup-checkbox-info": "Le mixage est une technique d'augmentation des données qui génère de nouveaux échantillons en mélangeant deux échantillons et leurs étiquettes.",
     "training-tab-use-mixup-checkbox-label": "Utiliser le mixage",
+    "training-tab-warning-detached-classifier-append": "Le mode d'ajout ne s'appliquera pas au classificateur détaché.",
     "validation-max-confidence-lower-than-min-confidence": "La confiance maximale doit être supérieure à la confiance minimale",
     "validation-no-audio-directory-selected": "Aucun dossiers audio sélectionnés",
     "validation-no-cache-file-selected": "Veuillez sélectionner un fichier cache.",
@@ -390,4 +392,4 @@
     "validation-no-valid-epoch-number": "Veuillez saisir un nombre valide d'époques.",
     "validation-no-valid-frequency": "Veuillez saisir une fréquence valide dans",
     "validation-no-valid-learning-rate": "Veuillez saisir un taux d'apprentissage valide."
-}
+}

birdnet_analyzer/lang/id.json

@@ -180,6 +180,8 @@
     "multi-tab-samples-dataframe-column-duration-header": "Panjang",
     "multi-tab-samples-dataframe-column-subpath-header": "Subjalur",
     "multi-tab-samples-dataframe-no-files-found": "Tidak ada file yang ditemukan",
+    "multi-tab-split-table-checkbox-info": "Kun tämä asetus on käytössä, jokaiselle syötetiedostolle luodaan oma tulostaulukko. Muussa tapauksessa kaikki tulokset yhdistetään yhteen taulukkoon.",
+    "multi-tab-split-table-checkbox-label": "Jaa taulukot syötetiedoston mukaan",
     "multi-tab-title": "Analisis batch",
     "progress-analyzing": "Menganalisis",
     "progress-autotune": "Autotune dalam progres",
@@ -369,13 +371,13 @@
     "training-tab-upsampling-radio-option-smote": "SMOTE",
     "training-tab-upsampling-ratio-slider-info": "Rasio minimum untuk kelas minoritas dibandingkan dengan kelas mayoritas setelah peningkatan sampel.",
     "training-tab-upsampling-ratio-slider-label": "Rasio peningkatan sampel",
-    "training-tab-warning-detached-classifier-append": "Mode tambah tidak akan berlaku untuk pengklasifikasi terlepas.",
     "training-tab-use-focal-loss-checkbox-info": "Menggunakan focal loss selama pelatihan. Ini memberikan bobot lebih pada contoh yang sulit dan membantu menangani ketidakseimbangan kelas.",
     "training-tab-use-focal-loss-checkbox-label": "Gunakan focal loss",
     "training-tab-use-labelsmoothing-checkbox-info": "Menerapkan label smoothing selama pelatihan. Ini dapat membantu meningkatkan generalisasi model dengan mencegah model menjadi terlalu percaya diri.",
     "training-tab-use-labelsmoothing-checkbox-label": "Gunakan label smoothing",
     "training-tab-use-mixup-checkbox-info": "Mixup adalah teknik augmentasi data yang menghasilkan sampel baru dengan mencampurkan dua sampel dan labelnya.",
     "training-tab-use-mixup-checkbox-label": "Gunakan mixup",
+    "training-tab-warning-detached-classifier-append": "Mode tambah tidak akan berlaku untuk pengklasifikasi terlepas.",
     "validation-max-confidence-lower-than-min-confidence": "Kepercayaan maksimum harus lebih besar dari kepercayaan minimum",
     "validation-no-audio-directory-selected": "Tidak ada audio direktori yang dipilih",
     "validation-no-cache-file-selected": "Silakan pilih berkas cache.",
@@ -390,4 +392,4 @@
     "validation-no-valid-epoch-number": "Silakan masukkan jumlah epochs yang valid.",
     "validation-no-valid-frequency": "Silakan masukkan frekuensi yang valid dalam",
     "validation-no-valid-learning-rate": "Silakan masukkan laju pembelajaran yang valid."
-}
+}