birdnet-analyzer 2.0.1__py3-none-any.whl → 2.1.0__py3-none-any.whl

birdnet_analyzer/analyze/__init__.py

@@ -1,5 +1,19 @@
+import os
+
+import birdnet_analyzer.config as cfg
 from birdnet_analyzer.analyze.core import analyze
 
+POSSIBLE_ADDITIONAL_COLUMNS_MAP = {
+    "lat": lambda: cfg.LATITUDE,
+    "lon": lambda: cfg.LONGITUDE,
+    "week": lambda: cfg.WEEK,
+    "overlap": lambda: cfg.SIG_OVERLAP,
+    "sensitivity": lambda: cfg.SIGMOID_SENSITIVITY,
+    "min_conf": lambda: cfg.MIN_CONFIDENCE,
+    "species_list": lambda: cfg.SPECIES_LIST_FILE or "",
+    "model": lambda: os.path.basename(cfg.MODEL_PATH),
+}
+
 __all__ = [
     "analyze",
 ]

birdnet_analyzer/analyze/cli.py

@@ -22,4 +22,9 @@ def main():
     except Exception:
         pass
 
+    if "additional_columns" in args and args.additional_columns and "csv" not in args.rtype:
+        import warnings
+
+        warnings.warn("The --additional_columns argument is only valid for CSV output. It will be ignored.", stacklevel=1)
+
     analyze(**vars(args))

birdnet_analyzer/analyze/core.py

@@ -26,6 +26,7 @@ def analyze(
     merge_consecutive: int = 1,
     threads: int = 8,
     locale: str = "en",
+    additional_columns: list[str] | None = None,
 ):
     """
     Analyzes audio files for bird species detection using the BirdNET-Analyzer.
@@ -53,6 +54,7 @@ def analyze(
         merge_consecutive (int, optional): Merge consecutive detections within this time window in seconds. Defaults to 1.
         threads (int, optional): Number of CPU threads to use for analysis. Defaults to 8.
         locale (str, optional): Locale for species names and output. Defaults to "en".
+        additional_columns (list[str] | None, optional): Additional columns to include in the output. Defaults to None.
     Returns:
         None
     Raises:
@@ -95,6 +97,7 @@ def analyze(
         skip_existing_results=skip_existing_results,
         threads=threads,
         labels_file=cfg.LABELS_FILE,
+        additional_columns=additional_columns,
     )
 
     print(f"Found {len(cfg.FILE_LIST)} files to analyze")
@@ -150,6 +153,7 @@ def _set_params(
     merge_consecutive,
     threads,
     labels_file=None,
+    additional_columns=None,
 ):
     import birdnet_analyzer.config as cfg
     from birdnet_analyzer.analyze.utils import load_codes
@@ -162,7 +166,7 @@ def _set_params(
     cfg.LOCATION_FILTER_THRESHOLD = sf_thresh
     cfg.TOP_N = top_n
     cfg.MERGE_CONSECUTIVE = merge_consecutive
-    cfg.INPUT_PATH = audio_input
+    cfg.INPUT_PATH = audio_input.replace("/", os.sep)
     cfg.MIN_CONFIDENCE = min_conf
     cfg.SIGMOID_SENSITIVITY = sensitivity
     cfg.SIG_OVERLAP = overlap
@@ -172,6 +176,7 @@ def _set_params(
     cfg.RESULT_TYPES = rtype
     cfg.COMBINE_RESULTS = combine_results
     cfg.BATCH_SIZE = bs
+    cfg.ADDITIONAL_COLUMNS = additional_columns
 
     if not output:
         if os.path.isfile(cfg.INPUT_PATH):

birdnet_analyzer/analyze/utils.py

@@ -4,17 +4,17 @@ import datetime
 import json
 import operator
 import os
+from collections.abc import Sequence
 
 import numpy as np
 
 import birdnet_analyzer.config as cfg
 from birdnet_analyzer import audio, model, utils
 
-RAVEN_TABLE_HEADER = "Selection\tView\tChannel\tBegin Time (s)\tEnd Time (s)\tLow Freq (Hz)\tHigh Freq (Hz)\tCommon Name\tSpecies Code\tConfidence\tBegin Path\tFile Offset (s)\n"  # noqa: E501
-RTABLE_HEADER = "filepath,start,end,scientific_name,common_name,confidence,lat,lon,week,overlap,sensitivity,min_conf,species_list,model\n"
-KALEIDOSCOPE_HEADER = (
-    "INDIR,FOLDER,IN FILE,OFFSET,DURATION,scientific_name,common_name,confidence,lat,lon,week,overlap,sensitivity\n"
+RAVEN_TABLE_HEADER = (
+    "Selection\tView\tChannel\tBegin Time (s)\tEnd Time (s)\tLow Freq (Hz)\tHigh Freq (Hz)\tCommon Name\tSpecies Code\tConfidence\tBegin Path\tFile Offset (s)\n"
 )
+KALEIDOSCOPE_HEADER = "INDIR,FOLDER,IN FILE,OFFSET,DURATION,scientific_name,common_name,confidence,lat,lon,week,overlap,sensitivity\n"
 CSV_HEADER = "Start (s),End (s),Scientific name,Common name,Confidence,File\n"
 SCRIPT_DIR = os.path.abspath(os.path.dirname(__file__))
 
@@ -80,7 +80,7 @@ def generate_raven_table(timestamps: list[str], result: dict[str, list], afile_p
 
     high_freq = min(high_freq, int(cfg.SIG_FMAX / cfg.AUDIO_SPEED))
 
-    high_freq = min(high_freq, int(cfg.BANDPASS_FMAX / cfg.AUDIO_SPEED))
+    high_freq = int(min(high_freq, int(cfg.BANDPASS_FMAX / cfg.AUDIO_SPEED)))
     low_freq = max(cfg.SIG_FMIN, int(cfg.BANDPASS_FMIN / cfg.AUDIO_SPEED))
 
     # Extract valid predictions for every timestamp
@@ -92,7 +92,9 @@ def generate_raven_table(timestamps: list[str], result: dict[str, list], afile_p
             selection_id += 1
             label = cfg.TRANSLATED_LABELS[cfg.LABELS.index(c[0])]
             code = cfg.CODES[c[0]] if c[0] in cfg.CODES else c[0]
-            rstring += f"{selection_id}\tSpectrogram 1\t1\t{start}\t{end}\t{low_freq}\t{high_freq}\t{label.split('_', 1)[-1]}\t{code}\t{c[1]:.4f}\t{afile_path}\t{start}\n"  # noqa: E501
+            rstring += (
+                f"{selection_id}\tSpectrogram 1\t1\t{start}\t{end}\t{low_freq}\t{high_freq}\t{label.split('_', 1)[-1]}\t{code}\t{c[1]:.4f}\t{afile_path}\t{start}\n"
+            )
 
         # Write result string to file
         out_string += rstring
@@ -103,9 +105,7 @@ def generate_raven_table(timestamps: list[str], result: dict[str, list], afile_p
     # of file durations during the analysis.
     if len(out_string) == len(RAVEN_TABLE_HEADER) and cfg.OUTPUT_PATH is not None:
         selection_id += 1
-        out_string += (
-            f"{selection_id}\tSpectrogram 1\t1\t0\t3\t{low_freq}\t{high_freq}\tnocall\tnocall\t1.0\t{afile_path}\t0\n"
-        )
+        out_string += f"{selection_id}\tSpectrogram 1\t1\t0\t3\t{low_freq}\t{high_freq}\tnocall\tnocall\t1.0\t{afile_path}\t0\n"
 
     utils.save_result_file(result_path, out_string)
 
@@ -202,7 +202,18 @@ def generate_csv(timestamps: list[str], result: dict[str, list], afile_path: str
     Returns:
         None
     """
+    from birdnet_analyzer.analyze import POSSIBLE_ADDITIONAL_COLUMNS_MAP
+
     out_string = CSV_HEADER
+    columns_map = {}
+
+    if cfg.ADDITIONAL_COLUMNS:
+        for col in cfg.ADDITIONAL_COLUMNS:
+            if col in POSSIBLE_ADDITIONAL_COLUMNS_MAP:
+                columns_map[col] = POSSIBLE_ADDITIONAL_COLUMNS_MAP[col]()
+
+        if columns_map:
+            out_string = out_string[:-1] + "," + ",".join(columns_map) + "\n"
 
     for timestamp in timestamps:
         rstring = ""
@@ -210,7 +221,12 @@ def generate_csv(timestamps: list[str], result: dict[str, list], afile_path: str
         for c in result[timestamp]:
             start, end = timestamp.split("-", 1)
             label = cfg.TRANSLATED_LABELS[cfg.LABELS.index(c[0])]
-            rstring += f"{start},{end},{label.split('_', 1)[0]},{label.split('_', 1)[-1]},{c[1]:.4f},{afile_path}\n"
+            rstring += f"{start},{end},{label.split('_', 1)[0]},{label.split('_', 1)[-1]},{c[1]:.4f},{afile_path}"
+
+            if columns_map:
+                rstring += "," + ",".join(str(val) for val in columns_map.values())
+
+            rstring += "\n"
 
         # Write result string to file
         out_string += rstring
@@ -364,29 +380,23 @@ def combine_csv_files(saved_results: list[str]):
     Args:
         saved_results (list[str]): A list of file paths to the CSV files to be combined.
     """
-    # Combine all files
-    with open(os.path.join(cfg.OUTPUT_PATH, cfg.OUTPUT_CSV_FILENAME), "w", encoding="utf-8") as f:
-        f.write(CSV_HEADER)
+    out_string = ""
 
-        for rfile in saved_results:
+    for rfile in saved_results:
+        try:
             with open(rfile, encoding="utf-8") as rf:
-                try:
-                    lines = rf.readlines()
+                lines = rf.readlines()
+                out_string += "".join(lines[1:] if out_string else lines)
 
-                    # make sure it's a selection table
-                    if "Start (s)" not in lines[0] or "Confidence" not in lines[0]:
-                        continue
-
-                    # skip header and add to file
-                    for line in lines[1:]:
-                        f.write(line)
+        except Exception as ex:
+            print(f"Error: Cannot combine results from {rfile}.\n", flush=True)
+            utils.write_error_log(ex)
 
-                except Exception as ex:
-                    print(f"Error: Cannot combine results from {rfile}.\n", flush=True)
-                    utils.write_error_log(ex)
+    with open(os.path.join(cfg.OUTPUT_PATH, cfg.OUTPUT_CSV_FILENAME), "w", encoding="utf-8") as f:
+        f.write(out_string)
 
 
-def combine_results(saved_results: list[dict[str, str]]):
+def combine_results(saved_results: Sequence[dict[str, str]| None]):
     """
     Combines various types of result files based on the configuration settings.
     This function checks the types of results specified in the configuration
@@ -403,9 +413,6 @@ def combine_results(saved_results: list[dict[str, str]]):
     if "table" in cfg.RESULT_TYPES:
         combine_raven_tables([f["table"] for f in saved_results if f])
 
-    # if "r" in cfg.RESULT_TYPES:
-    #     combine_rtable_files([f["r"] for f in saved_results if f])
-
     if "kaleidoscope" in cfg.RESULT_TYPES:
         combine_kaleidoscope_files([f["kaleidoscope"] for f in saved_results if f])
 
@@ -509,9 +516,7 @@ def get_raw_audio_from_file(fpath: str, offset, duration):
         The signal split into a list of chunks.
     """
     # Open file
-    sig, rate = audio.open_audio_file(
-        fpath, cfg.SAMPLE_RATE, offset, duration, cfg.BANDPASS_FMIN, cfg.BANDPASS_FMAX, cfg.AUDIO_SPEED
-    )
+    sig, rate = audio.open_audio_file(fpath, cfg.SAMPLE_RATE, offset, duration, cfg.BANDPASS_FMIN, cfg.BANDPASS_FMAX, cfg.AUDIO_SPEED)
 
     # Split into raw audio chunks
     return audio.split_signal(sig, rate, cfg.SIG_LENGTH, cfg.SIG_OVERLAP, cfg.SIG_MINLEN)
@@ -563,16 +568,14 @@ def get_result_file_names(fpath: str):
     # if "r" in cfg.RESULT_TYPES:
     #     result_names["r"] = os.path.join(cfg.OUTPUT_PATH, file_shorthand + ".BirdNET.results.r.csv")
     if "kaleidoscope" in cfg.RESULT_TYPES:
-        result_names["kaleidoscope"] = os.path.join(
-            cfg.OUTPUT_PATH, file_shorthand + ".BirdNET.results.kaleidoscope.csv"
-        )
+        result_names["kaleidoscope"] = os.path.join(cfg.OUTPUT_PATH, file_shorthand + ".BirdNET.results.kaleidoscope.csv")
     if "csv" in cfg.RESULT_TYPES:
         result_names["csv"] = os.path.join(cfg.OUTPUT_PATH, file_shorthand + ".BirdNET.results.csv")
 
     return result_names
 
 
-def analyze_file(item):
+def analyze_file(item) -> dict[str, str] | None:
     """
     Analyzes an audio file and generates prediction results.
 
@@ -606,7 +609,7 @@ def analyze_file(item):
     print(f"Analyzing {fpath}", flush=True)
 
     try:
-        fileLengthSeconds = int(audio.get_audio_file_length(fpath) / cfg.AUDIO_SPEED)
+        fileLengthSeconds = audio.get_audio_file_length(fpath)
     except Exception as ex:
         # Write error log
         print(f"Error: Cannot analyze audio file {fpath}. File corrupt?\n", flush=True)
@@ -649,8 +652,7 @@ def analyze_file(item):
                     p_labels = [
                         p
                         for p in zip(cfg.LABELS, pred, strict=True)
-                        if (cfg.TOP_N or p[1] >= cfg.MIN_CONFIDENCE)
-                        and (not cfg.SPECIES_LIST or p[0] in cfg.SPECIES_LIST)
+                        if (cfg.TOP_N or p[1] >= cfg.MIN_CONFIDENCE) and (not cfg.SPECIES_LIST or p[0] in cfg.SPECIES_LIST)
                     ]
 
                     # Sort by score

birdnet_analyzer/audio.py

@@ -61,7 +61,7 @@ def get_audio_file_length(path):
     """
     # Open file with librosa (uses ffmpeg or libav)
 
-    return librosa.get_duration(filename=path, sr=None)
+    return librosa.get_duration(path=path, sr=None)
 
 
 def get_sample_rate(path: str):
@@ -184,7 +184,7 @@ def split_signal(sig, rate, seconds, overlap, minlen, amount=None):
 
     # Split signal with overlap
     sig_splits = []
-    sig_splits.extend(data[i : i + chunksize] for i in range(0, lastchunkpos, stepsize))
+    sig_splits.extend(data[i : i + chunksize] for i in range(0, lastchunkpos + 1, stepsize))
 
     return sig_splits
 

birdnet_analyzer/cli.py

@@ -31,7 +31,7 @@ ASCII_LOGO = r"""
                                                  **=====        
                                                ***+==           
                                               ****+             
-""" # noqa: W291
+"""  # noqa: W291
 
 
 def io_args():
@@ -80,10 +80,9 @@ def bandpass_args():
 
     return p
 
-
-def species_args():
+def species_list_args():
     """
-    Creates an argument parser for species-related arguments.
+    Creates an argument parser for species-list arguments.
     Returns:
         argparse.ArgumentParser: The argument parser with the following arguments:
             --lat (float): Recording location latitude. Set -1 to ignore. Default is -1.
@@ -104,17 +103,33 @@ def species_args():
         default=-1,
         help="Week of the year when the recording was made. Values in [1, 48] (4 weeks per month). Set -1 for year-round species list.",
     )
-    p.add_argument(
-        "--slist",
-        help='Path to species list file or folder. If folder is provided, species list needs to be named "species_list.txt". If lat and lon are provided, this list will be ignored.',
-    )
     p.add_argument(
         "--sf_thresh",
         type=lambda a: max(0.0001, min(0.99, float(a))),
         default=cfg.LOCATION_FILTER_THRESHOLD,
         help="Minimum species occurrence frequency threshold for location filter. Values in [0.0001, 0.99].",
     )
+    return p
 
+def species_args():
+    """
+    Creates an argument parser for species-related arguments including the species-list arguments.
+    Returns:
+        argparse.ArgumentParser: The argument parser with the following arguments:
+            --lat (float): Recording location latitude. Set -1 to ignore. Default is -1.
+            --lon (float): Recording location longitude. Set -1 to ignore. Default is -1.
+            --week (int): Week of the year when the recording was made. Values in [1, 48] (4 weeks per month).
+                          Set -1 for year-round species list. Default is -1.
+            --sf_thresh (float): Minimum species occurrence frequency threshold for location filter. Values in [0.01, 0.99].
+                                 Defaults to cfg.LOCATION_FILTER_THRESHOLD.
+            --slist (str): Path to species list file or folder. If folder is provided, species list needs to be named
+                           "species_list.txt". If lat and lon are provided, this list will be ignored.
+    """
+    p = species_list_args()
+    p.add_argument(
+        "--slist",
+        help='Path to species list file or folder. If folder is provided, species list needs to be named "species_list.txt". If lat and lon are provided, this list will be ignored.',
+    )
     return p
 
 
@@ -309,6 +324,7 @@ def analyzer_parser():
     Returns:
         argparse.ArgumentParser: Configured argument parser for the BirdNET Analyzer CLI.
     """
+    from birdnet_analyzer.analyze import POSSIBLE_ADDITIONAL_COLUMNS_MAP
     parents = [
         io_args(),
         bandpass_args(),
@@ -339,6 +355,14 @@ def analyzer_parser():
         help="Specifies output format. Values in `['table', 'audacity',  'kaleidoscope', 'csv']`.",
         action=UniqueSetAction,
     )
+    parser.add_argument(
+        "--additional_columns",
+        choices=POSSIBLE_ADDITIONAL_COLUMNS_MAP.keys(),
+        nargs="+",
+        help="Additional columns to include in the output, only applied to the csv output format. "
+        + f"Values in [{','.join(POSSIBLE_ADDITIONAL_COLUMNS_MAP.keys())}].",
+        action=UniqueSetAction,
+    )
     parser.add_argument(
         "--combine_results",
         help="Also outputs a combined file for all the selected result types. If not set combined tables will be generated.",
@@ -404,6 +428,10 @@ def embeddings_parser():
         help="Path to input file or folder.",
     )
 
+    parser.add_argument(
+        "--file_output",
+    )
+
     return parser
 
 
@@ -498,9 +526,7 @@ def segments_parser():
     )
     parser.add_argument("audio_input", metavar="INPUT", help="Path to folder containing audio files.")
     parser.add_argument("-r", "--results", help="Path to folder containing result files. Defaults to the `input` path.")
-    parser.add_argument(
-        "-o", "--output", help="Output folder path for extracted segments. Defaults to the `input` path."
-    )
+    parser.add_argument("-o", "--output", help="Output folder path for extracted segments. Defaults to the `input` path.")
     parser.add_argument(
         "--max_segments",
         type=lambda a: max(1, int(a)),
@@ -535,9 +561,7 @@ def server_parser():
     parser.add_argument("-p", "--port", type=int, default=8080, help="Port of API endpoint server.")
     parser.add_argument(
         "--spath",
-        default="uploads/"
-        if os.environ.get("IS_GITHUB_RUNNER", "false").lower() == "true"
-        else os.path.join(SCRIPT_DIR, "uploads"),
+        default="uploads/" if os.environ.get("IS_GITHUB_RUNNER", "false").lower() == "true" else os.path.join(SCRIPT_DIR, "uploads"),
         help="Path to folder where uploaded files should be stored.",
     )
 
@@ -555,7 +579,7 @@ def species_parser():
     """
     parser = argparse.ArgumentParser(
         formatter_class=argparse.ArgumentDefaultsHelpFormatter,
-        parents=[species_args()],
+        parents=[species_list_args()],
     )
     parser.add_argument(
         "output",
@@ -603,9 +627,7 @@ def train_parser():
         metavar="INPUT",
         help="Path to training data folder. Subfolder names are used as labels.",
     )
-    parser.add_argument(
-        "--test_data", help="Path to test data folder. If not specified, a random validation split will be used."
-    )
+    parser.add_argument("--test_data", help="Path to test data folder. If not specified, a random validation split will be used.")
     parser.add_argument(
         "--crop_mode",
         default=cfg.SAMPLE_CROP_MODE,
@@ -633,6 +655,7 @@ def train_parser():
     )
     parser.add_argument(
         "--focal-loss",
+        dest="use_focal_loss",
         action="store_true",
         help="Use focal loss for training (helps with imbalanced classes and hard examples).",
     )

birdnet_analyzer/config.py

@@ -69,8 +69,8 @@ LOCATION_FILTER_THRESHOLD: float = 0.03
 
 # If None or empty file, no custom species list will be used
 # Note: Entries in this list have to match entries from the LABELS_FILE
-# We use the 2021 eBird taxonomy for species names (Clements list)
-CODES_FILE: str = os.path.join(SCRIPT_DIR, "eBird_taxonomy_codes_2021E.json")
+# We use the 2024 eBird taxonomy for species names (Clements list)
+CODES_FILE: str = os.path.join(SCRIPT_DIR, "eBird_taxonomy_codes_2024E.json")
 SPECIES_LIST_FILE: str = os.path.join(SCRIPT_DIR, "example/species_list.txt")
 
 # Supported file types
@@ -114,6 +114,7 @@ USE_NOISE: bool = False
 # 'audacity' denotes a TXT file with the same format as Audacity timeline labels
 # 'csv' denotes a generic CSV file with start, end, species and confidence.
 RESULT_TYPES: set[str] | list[str] = {"table"}
+ADDITIONAL_COLUMNS: list[str] | None = None
 OUTPUT_RAVEN_FILENAME: str = "BirdNET_SelectionTable.txt"  # this is for combined Raven selection tables only
 # OUTPUT_RTABLE_FILENAME: str = "BirdNET_RTable.csv"
 OUTPUT_KALEIDOSCOPE_FILENAME: str = "BirdNET_Kaleidoscope.csv"
@@ -226,7 +227,7 @@ FILE_STORAGE_PATH: str = ""
 # Path to custom trained classifier
 # If None, no custom classifier will be used
 # Make sure to set the LABELS_FILE above accordingly
-CUSTOM_CLASSIFIER = None
+CUSTOM_CLASSIFIER: str | None = None
 
 ######################
 # Get and set config #