Spaces:
Running
Running
BilalSardar
commited on
Commit
•
9d7f3c9
1
Parent(s):
a28423c
Update app.py
Browse files
app.py
CHANGED
@@ -84,54 +84,6 @@ def runSingleFileAnalysis(
|
|
84 |
)
|
85 |
|
86 |
|
87 |
-
def runBatchAnalysis(
|
88 |
-
output_path,
|
89 |
-
confidence,
|
90 |
-
sensitivity,
|
91 |
-
overlap,
|
92 |
-
species_list_choice,
|
93 |
-
species_list_file,
|
94 |
-
lat,
|
95 |
-
lon,
|
96 |
-
week,
|
97 |
-
use_yearlong,
|
98 |
-
sf_thresh,
|
99 |
-
custom_classifier_file,
|
100 |
-
output_type,
|
101 |
-
locale,
|
102 |
-
batch_size,
|
103 |
-
threads,
|
104 |
-
input_dir,
|
105 |
-
progress=gr.Progress(),
|
106 |
-
):
|
107 |
-
validate(input_dir, "Please select a directory.")
|
108 |
-
batch_size = int(batch_size)
|
109 |
-
threads = int(threads)
|
110 |
-
|
111 |
-
if species_list_choice == _CUSTOM_SPECIES:
|
112 |
-
validate(species_list_file, "Please select a species list.")
|
113 |
-
|
114 |
-
return runAnalysis(
|
115 |
-
None,
|
116 |
-
output_path,
|
117 |
-
confidence,
|
118 |
-
sensitivity,
|
119 |
-
overlap,
|
120 |
-
species_list_choice,
|
121 |
-
species_list_file,
|
122 |
-
lat,
|
123 |
-
lon,
|
124 |
-
week,
|
125 |
-
use_yearlong,
|
126 |
-
sf_thresh,
|
127 |
-
custom_classifier_file,
|
128 |
-
output_type,
|
129 |
-
"en" if not locale else locale,
|
130 |
-
batch_size if batch_size and batch_size > 0 else 1,
|
131 |
-
threads if threads and threads > 0 else 4,
|
132 |
-
input_dir,
|
133 |
-
progress,
|
134 |
-
)
|
135 |
|
136 |
|
137 |
def runAnalysis(
|
@@ -431,147 +383,6 @@ def select_directory(collect_files=True):
|
|
431 |
return dir_name[0] if dir_name else None
|
432 |
|
433 |
|
434 |
-
def start_training(
|
435 |
-
data_dir,
|
436 |
-
crop_mode,
|
437 |
-
crop_overlap,
|
438 |
-
output_dir,
|
439 |
-
classifier_name,
|
440 |
-
epochs,
|
441 |
-
batch_size,
|
442 |
-
learning_rate,
|
443 |
-
hidden_units,
|
444 |
-
use_mixup,
|
445 |
-
upsampling_ratio,
|
446 |
-
upsampling_mode,
|
447 |
-
model_format,
|
448 |
-
progress=gr.Progress(),
|
449 |
-
):
|
450 |
-
"""Starts the training of a custom classifier.
|
451 |
-
|
452 |
-
Args:
|
453 |
-
data_dir: Directory containing the training data.
|
454 |
-
output_dir: Directory for the new classifier.
|
455 |
-
classifier_name: File name of the classifier.
|
456 |
-
epochs: Number of epochs to train for.
|
457 |
-
batch_size: Number of samples in one batch.
|
458 |
-
learning_rate: Learning rate for training.
|
459 |
-
hidden_units: If > 0 the classifier contains a further hidden layer.
|
460 |
-
progress: The gradio progress bar.
|
461 |
-
|
462 |
-
Returns:
|
463 |
-
Returns a matplotlib.pyplot figure.
|
464 |
-
"""
|
465 |
-
validate(data_dir, "Please select your Training data.")
|
466 |
-
validate(output_dir, "Please select a directory for the classifier.")
|
467 |
-
validate(classifier_name, "Please enter a valid name for the classifier.")
|
468 |
-
|
469 |
-
if not epochs or epochs < 0:
|
470 |
-
raise gr.Error("Please enter a valid number of epochs.")
|
471 |
-
|
472 |
-
if not batch_size or batch_size < 0:
|
473 |
-
raise gr.Error("Please enter a valid batch size.")
|
474 |
-
|
475 |
-
if not learning_rate or learning_rate < 0:
|
476 |
-
raise gr.Error("Please enter a valid learning rate.")
|
477 |
-
|
478 |
-
if not hidden_units or hidden_units < 0:
|
479 |
-
hidden_units = 0
|
480 |
-
|
481 |
-
if progress is not None:
|
482 |
-
progress((0, epochs), desc="Loading data & building classifier", unit="epoch")
|
483 |
-
|
484 |
-
cfg.TRAIN_DATA_PATH = data_dir
|
485 |
-
cfg.SAMPLE_CROP_MODE = crop_mode
|
486 |
-
cfg.SIG_OVERLAP = crop_overlap
|
487 |
-
cfg.CUSTOM_CLASSIFIER = str(Path(output_dir) / classifier_name)
|
488 |
-
cfg.TRAIN_EPOCHS = int(epochs)
|
489 |
-
cfg.TRAIN_BATCH_SIZE = int(batch_size)
|
490 |
-
cfg.TRAIN_LEARNING_RATE = learning_rate
|
491 |
-
cfg.TRAIN_HIDDEN_UNITS = int(hidden_units)
|
492 |
-
cfg.TRAIN_WITH_MIXUP = use_mixup
|
493 |
-
cfg.UPSAMPLING_RATIO = min(max(0, upsampling_ratio), 1)
|
494 |
-
cfg.UPSAMPLING_MODE = upsampling_mode
|
495 |
-
cfg.TRAINED_MODEL_OUTPUT_FORMAT = model_format
|
496 |
-
|
497 |
-
def progression(epoch, logs=None):
|
498 |
-
if progress is not None:
|
499 |
-
if epoch + 1 == epochs:
|
500 |
-
progress((epoch + 1, epochs), total=epochs, unit="epoch", desc=f"Saving at {cfg.CUSTOM_CLASSIFIER}")
|
501 |
-
else:
|
502 |
-
progress((epoch + 1, epochs), total=epochs, unit="epoch")
|
503 |
-
|
504 |
-
history = trainModel(on_epoch_end=progression)
|
505 |
-
|
506 |
-
if len(history.epoch) < epochs:
|
507 |
-
gr.Info("Stopped early - validation metric not improving.")
|
508 |
-
|
509 |
-
auprc = history.history["val_AUPRC"]
|
510 |
-
|
511 |
-
import matplotlib.pyplot as plt
|
512 |
-
|
513 |
-
fig = plt.figure()
|
514 |
-
plt.plot(auprc)
|
515 |
-
plt.ylabel("Area under precision-recall curve")
|
516 |
-
plt.xlabel("Epoch")
|
517 |
-
|
518 |
-
return fig
|
519 |
-
|
520 |
-
|
521 |
-
def extract_segments(audio_dir, result_dir, output_dir, min_conf, num_seq, seq_length, threads, progress=gr.Progress()):
|
522 |
-
validate(audio_dir, "No audio directory selected")
|
523 |
-
|
524 |
-
if not result_dir:
|
525 |
-
result_dir = audio_dir
|
526 |
-
|
527 |
-
if not output_dir:
|
528 |
-
output_dir = audio_dir
|
529 |
-
|
530 |
-
if progress is not None:
|
531 |
-
progress(0, desc="Searching files ...")
|
532 |
-
|
533 |
-
# Parse audio and result folders
|
534 |
-
cfg.FILE_LIST = segments.parseFolders(audio_dir, result_dir)
|
535 |
-
|
536 |
-
# Set output folder
|
537 |
-
cfg.OUTPUT_PATH = output_dir
|
538 |
-
|
539 |
-
# Set number of threads
|
540 |
-
cfg.CPU_THREADS = int(threads)
|
541 |
-
|
542 |
-
# Set confidence threshold
|
543 |
-
cfg.MIN_CONFIDENCE = max(0.01, min(0.99, min_conf))
|
544 |
-
|
545 |
-
# Parse file list and make list of segments
|
546 |
-
cfg.FILE_LIST = segments.parseFiles(cfg.FILE_LIST, max(1, int(num_seq)))
|
547 |
-
|
548 |
-
# Add config items to each file list entry.
|
549 |
-
# We have to do this for Windows which does not
|
550 |
-
# support fork() and thus each process has to
|
551 |
-
# have its own config. USE LINUX!
|
552 |
-
flist = [(entry, max(cfg.SIG_LENGTH, float(seq_length)), cfg.getConfig()) for entry in cfg.FILE_LIST]
|
553 |
-
|
554 |
-
result_list = []
|
555 |
-
|
556 |
-
# Extract segments
|
557 |
-
if cfg.CPU_THREADS < 2:
|
558 |
-
for i, entry in enumerate(flist):
|
559 |
-
result = extractSegments_wrapper(entry)
|
560 |
-
result_list.append(result)
|
561 |
-
|
562 |
-
if progress is not None:
|
563 |
-
progress((i, len(flist)), total=len(flist), unit="files")
|
564 |
-
else:
|
565 |
-
with concurrent.futures.ProcessPoolExecutor(max_workers=cfg.CPU_THREADS) as executor:
|
566 |
-
futures = (executor.submit(extractSegments_wrapper, arg) for arg in flist)
|
567 |
-
for i, f in enumerate(concurrent.futures.as_completed(futures), start=1):
|
568 |
-
if progress is not None:
|
569 |
-
progress((i, len(flist)), total=len(flist), unit="files")
|
570 |
-
result = f.result()
|
571 |
-
|
572 |
-
result_list.append(result)
|
573 |
-
|
574 |
-
return [[os.path.relpath(r[0], audio_dir), r[1]] for r in result_list]
|
575 |
|
576 |
|
577 |
def sample_sliders(opened=True):
|
@@ -761,269 +572,6 @@ if __name__ == "__main__":
|
|
761 |
|
762 |
single_file_analyze.click(runSingleFileAnalysis, inputs=inputs, outputs=output_dataframe)
|
763 |
|
764 |
-
def build_multi_analysis_tab():
|
765 |
-
with gr.Tab("Multiple files"):
|
766 |
-
input_directory_state = gr.State()
|
767 |
-
output_directory_predict_state = gr.State()
|
768 |
-
with gr.Row():
|
769 |
-
with gr.Column():
|
770 |
-
select_directory_btn = gr.Button("Select directory (recursive)")
|
771 |
-
directory_input = gr.Matrix(interactive=False, elem_classes="mh-200", headers=["Subpath", "Length"])
|
772 |
-
|
773 |
-
def select_directory_on_empty():
|
774 |
-
res = select_directory()
|
775 |
-
|
776 |
-
return res if res[1] else [res[0], [["No files found"]]]
|
777 |
-
|
778 |
-
select_directory_btn.click(
|
779 |
-
select_directory_on_empty, outputs=[input_directory_state, directory_input], show_progress=True
|
780 |
-
)
|
781 |
-
|
782 |
-
with gr.Column():
|
783 |
-
select_out_directory_btn = gr.Button("Select output directory.")
|
784 |
-
selected_out_textbox = gr.Textbox(
|
785 |
-
label="Output directory",
|
786 |
-
interactive=False,
|
787 |
-
placeholder="If not selected, the input directory will be used.",
|
788 |
-
)
|
789 |
-
|
790 |
-
def select_directory_wrapper():
|
791 |
-
return (select_directory(collect_files=False),) * 2
|
792 |
-
|
793 |
-
select_out_directory_btn.click(
|
794 |
-
select_directory_wrapper,
|
795 |
-
outputs=[output_directory_predict_state, selected_out_textbox],
|
796 |
-
show_progress=False,
|
797 |
-
)
|
798 |
-
|
799 |
-
confidence_slider, sensitivity_slider, overlap_slider = sample_sliders()
|
800 |
-
|
801 |
-
(
|
802 |
-
species_list_radio,
|
803 |
-
species_file_input,
|
804 |
-
lat_number,
|
805 |
-
lon_number,
|
806 |
-
week_number,
|
807 |
-
sf_thresh_number,
|
808 |
-
yearlong_checkbox,
|
809 |
-
selected_classifier_state,
|
810 |
-
) = species_lists()
|
811 |
-
|
812 |
-
output_type_radio = gr.Radio(
|
813 |
-
list(OUTPUT_TYPE_MAP.keys()),
|
814 |
-
value="Raven selection table",
|
815 |
-
label="Result type",
|
816 |
-
info="Specifies output format.",
|
817 |
-
)
|
818 |
-
|
819 |
-
with gr.Row():
|
820 |
-
batch_size_number = gr.Number(
|
821 |
-
precision=1, label="Batch size", value=1, info="Number of samples to process at the same time."
|
822 |
-
)
|
823 |
-
threads_number = gr.Number(precision=1, label="Threads", value=4, info="Number of CPU threads.")
|
824 |
-
|
825 |
-
locale_radio = locale()
|
826 |
-
|
827 |
-
start_batch_analysis_btn = gr.Button("Analyze")
|
828 |
-
|
829 |
-
result_grid = gr.Matrix(headers=["File", "Execution"], elem_classes="mh-200")
|
830 |
-
|
831 |
-
inputs = [
|
832 |
-
output_directory_predict_state,
|
833 |
-
confidence_slider,
|
834 |
-
sensitivity_slider,
|
835 |
-
overlap_slider,
|
836 |
-
species_list_radio,
|
837 |
-
species_file_input,
|
838 |
-
lat_number,
|
839 |
-
lon_number,
|
840 |
-
week_number,
|
841 |
-
yearlong_checkbox,
|
842 |
-
sf_thresh_number,
|
843 |
-
selected_classifier_state,
|
844 |
-
output_type_radio,
|
845 |
-
locale_radio,
|
846 |
-
batch_size_number,
|
847 |
-
threads_number,
|
848 |
-
input_directory_state,
|
849 |
-
]
|
850 |
-
|
851 |
-
start_batch_analysis_btn.click(runBatchAnalysis, inputs=inputs, outputs=result_grid)
|
852 |
-
|
853 |
-
def build_train_tab():
|
854 |
-
with gr.Tab("Train"):
|
855 |
-
input_directory_state = gr.State()
|
856 |
-
output_directory_state = gr.State()
|
857 |
-
|
858 |
-
with gr.Row():
|
859 |
-
with gr.Column():
|
860 |
-
select_directory_btn = gr.Button("Training data")
|
861 |
-
directory_input = gr.List(headers=["Classes"], interactive=False, elem_classes="mh-200")
|
862 |
-
select_directory_btn.click(
|
863 |
-
select_subdirectories, outputs=[input_directory_state, directory_input], show_progress=False
|
864 |
-
)
|
865 |
-
|
866 |
-
with gr.Column():
|
867 |
-
select_directory_btn = gr.Button("Classifier output")
|
868 |
-
|
869 |
-
with gr.Column():
|
870 |
-
classifier_name = gr.Textbox(
|
871 |
-
"CustomClassifier",
|
872 |
-
visible=False,
|
873 |
-
info="The name of the new classifier.",
|
874 |
-
)
|
875 |
-
output_format = gr.Radio(
|
876 |
-
["tflite", "raven", "both"],
|
877 |
-
value="tflite",
|
878 |
-
label="Model output format",
|
879 |
-
info="Format for the trained classifier.",
|
880 |
-
visible=False,
|
881 |
-
)
|
882 |
-
|
883 |
-
def select_directory_and_update_tb():
|
884 |
-
dir_name = _WINDOW.create_file_dialog(webview.FOLDER_DIALOG)
|
885 |
-
|
886 |
-
if dir_name:
|
887 |
-
return (
|
888 |
-
dir_name[0],
|
889 |
-
gr.Textbox.update(label=dir_name[0] + "\\", visible=True),
|
890 |
-
gr.Radio.update(visible=True, interactive=True),
|
891 |
-
)
|
892 |
-
|
893 |
-
return None, None
|
894 |
-
|
895 |
-
select_directory_btn.click(
|
896 |
-
select_directory_and_update_tb,
|
897 |
-
outputs=[output_directory_state, classifier_name, output_format],
|
898 |
-
show_progress=False,
|
899 |
-
)
|
900 |
-
|
901 |
-
with gr.Row():
|
902 |
-
epoch_number = gr.Number(100, label="Epochs", info="Number of training epochs.")
|
903 |
-
batch_size_number = gr.Number(32, label="Batch size", info="Batch size.")
|
904 |
-
learning_rate_number = gr.Number(0.01, label="Learning rate", info="Learning rate.")
|
905 |
-
|
906 |
-
with gr.Row():
|
907 |
-
crop_mode = gr.Radio(
|
908 |
-
["center", "first", "segments"],
|
909 |
-
value="center",
|
910 |
-
label="Crop mode",
|
911 |
-
info="Crop mode for training data.",
|
912 |
-
)
|
913 |
-
crop_overlap = gr.Number(0.0, label="Crop overlap", info="Overlap of training data segments", visible=False)
|
914 |
-
|
915 |
-
def on_crop_select(new_crop_mode):
|
916 |
-
return gr.Number.update(visible=new_crop_mode == "segments", interactive=new_crop_mode == "segments")
|
917 |
-
|
918 |
-
crop_mode.change(on_crop_select, inputs=crop_mode, outputs=crop_overlap)
|
919 |
-
|
920 |
-
with gr.Row():
|
921 |
-
upsampling_mode = gr.Radio(
|
922 |
-
["repeat", "mean", "smote"],
|
923 |
-
value="repeat",
|
924 |
-
label="Upsampling mode",
|
925 |
-
info="Balance data through upsampling.",
|
926 |
-
)
|
927 |
-
upsampling_ratio = gr.Slider(
|
928 |
-
0.0, 1.0, 0.0, step=0.01, label="Upsampling ratio", info="Balance train data and upsample minority classes."
|
929 |
-
)
|
930 |
-
|
931 |
-
with gr.Row():
|
932 |
-
hidden_units_number = gr.Number(
|
933 |
-
0, label="Hidden units", info="Number of hidden units. If set to >0, a two-layer classifier is used."
|
934 |
-
)
|
935 |
-
use_mixup = gr.Checkbox(False, label="Use mixup", info="Whether to use mixup for training.", show_label=True)
|
936 |
-
|
937 |
-
train_history_plot = gr.Plot()
|
938 |
-
|
939 |
-
start_training_button = gr.Button("Start training")
|
940 |
-
|
941 |
-
start_training_button.click(
|
942 |
-
start_training,
|
943 |
-
inputs=[
|
944 |
-
input_directory_state,
|
945 |
-
crop_mode,
|
946 |
-
crop_overlap,
|
947 |
-
output_directory_state,
|
948 |
-
classifier_name,
|
949 |
-
epoch_number,
|
950 |
-
batch_size_number,
|
951 |
-
learning_rate_number,
|
952 |
-
hidden_units_number,
|
953 |
-
use_mixup,
|
954 |
-
upsampling_ratio,
|
955 |
-
upsampling_mode,
|
956 |
-
output_format,
|
957 |
-
],
|
958 |
-
outputs=[train_history_plot],
|
959 |
-
)
|
960 |
-
|
961 |
-
def build_segments_tab():
|
962 |
-
with gr.Tab("Segments"):
|
963 |
-
audio_directory_state = gr.State()
|
964 |
-
result_directory_state = gr.State()
|
965 |
-
output_directory_state = gr.State()
|
966 |
-
|
967 |
-
def select_directory_to_state_and_tb():
|
968 |
-
return (select_directory(collect_files=False),) * 2
|
969 |
-
|
970 |
-
with gr.Row():
|
971 |
-
select_audio_directory_btn = gr.Button("Select audio directory (recursive)")
|
972 |
-
selected_audio_directory_tb = gr.Textbox(show_label=False, interactive=False)
|
973 |
-
select_audio_directory_btn.click(
|
974 |
-
select_directory_to_state_and_tb,
|
975 |
-
outputs=[selected_audio_directory_tb, audio_directory_state],
|
976 |
-
show_progress=False,
|
977 |
-
)
|
978 |
-
|
979 |
-
with gr.Row():
|
980 |
-
select_result_directory_btn = gr.Button("Select result directory")
|
981 |
-
selected_result_directory_tb = gr.Textbox(
|
982 |
-
show_label=False, interactive=False, placeholder="Same as audio directory if not selected"
|
983 |
-
)
|
984 |
-
select_result_directory_btn.click(
|
985 |
-
select_directory_to_state_and_tb,
|
986 |
-
outputs=[result_directory_state, selected_result_directory_tb],
|
987 |
-
show_progress=False,
|
988 |
-
)
|
989 |
-
|
990 |
-
with gr.Row():
|
991 |
-
select_output_directory_btn = gr.Button("Select output directory")
|
992 |
-
selected_output_directory_tb = gr.Textbox(
|
993 |
-
show_label=False, interactive=False, placeholder="Same as audio directory if not selected"
|
994 |
-
)
|
995 |
-
select_output_directory_btn.click(
|
996 |
-
select_directory_to_state_and_tb,
|
997 |
-
outputs=[selected_output_directory_tb, output_directory_state],
|
998 |
-
show_progress=False,
|
999 |
-
)
|
1000 |
-
|
1001 |
-
min_conf_slider = gr.Slider(
|
1002 |
-
minimum=0.1, maximum=0.99, step=0.01, label="Minimum confidence", info="Minimum confidence threshold."
|
1003 |
-
)
|
1004 |
-
num_seq_number = gr.Number(
|
1005 |
-
100, label="Max number of segments", info="Maximum number of randomly extracted segments per species."
|
1006 |
-
)
|
1007 |
-
seq_length_number = gr.Number(3.0, label="Sequence length", info="Length of extracted segments in seconds.")
|
1008 |
-
threads_number = gr.Number(4, label="Threads", info="Number of CPU threads.")
|
1009 |
-
|
1010 |
-
extract_segments_btn = gr.Button("Extract segments")
|
1011 |
-
|
1012 |
-
result_grid = gr.Matrix(headers=["File", "Execution"], elem_classes="mh-200")
|
1013 |
-
|
1014 |
-
extract_segments_btn.click(
|
1015 |
-
extract_segments,
|
1016 |
-
inputs=[
|
1017 |
-
audio_directory_state,
|
1018 |
-
result_directory_state,
|
1019 |
-
output_directory_state,
|
1020 |
-
min_conf_slider,
|
1021 |
-
num_seq_number,
|
1022 |
-
seq_length_number,
|
1023 |
-
threads_number,
|
1024 |
-
],
|
1025 |
-
outputs=result_grid,
|
1026 |
-
)
|
1027 |
|
1028 |
with gr.Blocks(
|
1029 |
css=r".d-block .wrap {display: block !important;} .mh-200 {max-height: 300px; overflow-y: auto !important;} footer {display: none !important;} #single_file_audio, #single_file_audio * {max-height: 81.6px; min-height: 0;}",
|
@@ -1031,11 +579,8 @@ if __name__ == "__main__":
|
|
1031 |
analytics_enabled=False,
|
1032 |
) as demo:
|
1033 |
build_single_analysis_tab()
|
1034 |
-
#build_multi_analysis_tab()
|
1035 |
-
#build_train_tab()
|
1036 |
-
# build_segments_tab()
|
1037 |
|
1038 |
-
demo.launch()
|
1039 |
#url = demo.queue(api_open=False).launch(prevent_thread_lock=True, quiet=True)[1]
|
1040 |
#_WINDOW = webview.create_window("BirdNET-Analyzer", url.rstrip("/") + "?__theme=light", min_size=(1024, 768))
|
1041 |
|
|
|
84 |
)
|
85 |
|
86 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
87 |
|
88 |
|
89 |
def runAnalysis(
|
|
|
383 |
return dir_name[0] if dir_name else None
|
384 |
|
385 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
386 |
|
387 |
|
388 |
def sample_sliders(opened=True):
|
|
|
572 |
|
573 |
single_file_analyze.click(runSingleFileAnalysis, inputs=inputs, outputs=output_dataframe)
|
574 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
575 |
|
576 |
with gr.Blocks(
|
577 |
css=r".d-block .wrap {display: block !important;} .mh-200 {max-height: 300px; overflow-y: auto !important;} footer {display: none !important;} #single_file_audio, #single_file_audio * {max-height: 81.6px; min-height: 0;}",
|
|
|
579 |
analytics_enabled=False,
|
580 |
) as demo:
|
581 |
build_single_analysis_tab()
|
|
|
|
|
|
|
582 |
|
583 |
+
demo.launch(show_api=True)
|
584 |
#url = demo.queue(api_open=False).launch(prevent_thread_lock=True, quiet=True)[1]
|
585 |
#_WINDOW = webview.create_window("BirdNET-Analyzer", url.rstrip("/") + "?__theme=light", min_size=(1024, 768))
|
586 |
|