{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":91844,"databundleVersionId":11361821,"sourceType":"competition"}],"dockerImageVersionId":30918,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# 🐦 BirdCLEF+ EDA Tutorial\n---\n## 👀 Overview\nHi! I'm Coz, here to step you through a basic Exploratory Data Analysis (EDA) for the BirdCLEF+ 2025 Competition! This is an important step in understanding the problem before beginning work building machine learning models.\n\nHere I'll take you through how to explore some of the data that's been provided and create useful visualisations. Please note that some of the interactive components may require you to manually run the cell first in order to work.\n\nRemember to leave an upvote and a comment if you like it so that other people can find it too!","metadata":{}},{"cell_type":"markdown","source":"### General Info\nThe competition organisers have recorded a bunch of audio clips from El Silencio Natural Reserve, Colombia, and they want to know what animals they can find in them. The problem is, they don't want to show us any examples at all. So we have to design an algorithm which can work on pretty much any audio sample, pretty tough!\n\nTo help us, the organisers have assembled some publicly available audio samples from around the world but emphasising near Colombia. We can look at these to hopefully train a good model.\n\n","metadata":{}},{"cell_type":"markdown","source":"### Scoring\nThis competition is scored using ROC AUC (Area Under the Receiver Operating Characteristic Curve).\n\nThe solutions have True or False values for each species which may or may not be in a given audio sample. Our job is to decide on a probability that the animal is in the audio sample or not.\n\nThe ROC curve describes how the True Positive Rate changes as a function of the False Positive Rate, while we change a threshold value. The threshold value will determine whether our probability prediction like 0.8 is classed as a Positive prediction or Negative prediction. Here's a nice resource to explain it further - https://sefiks.com/2020/12/10/a-gentle-introduction-to-roc-curve-and-auc/\n\n![image.png](attachment:aef7ff13-c939-4044-ac0f-e5672017507d.png)\n\n'Macro' ROC AUC is a version of this algorithm where for each class (aka species) the ROC AUC is calculated, then these numbers are averaged across the classes. So we have to achieve good scores for all the animals, even the ones we have very little data for.","metadata":{},"attachments":{"aef7ff13-c939-4044-ac0f-e5672017507d.png":{"image/png":"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"},"d6557c79-0f96-4213-b29b-b9737d175d6e.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"### Aims\n\nWe want to identify which species are present in the audio clip with as much accuracy as possible.","metadata":{}},{"cell_type":"markdown","source":"---\n# ⚙️ Setup\nFirst we need to import the necessary libraries, set the right settings and prepare the right tools for our EDA.","metadata":{}},{"cell_type":"code","source":"# Essential Imports\nimport numpy as np\nimport pandas as pd\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nfrom tqdm import tqdm\nimport os\nimport ipywidgets as widgets\nfrom IPython.display import display\nimport warnings\n\n# Niche Imports\nimport IPython\nimport folium\nfrom folium.plugins import FastMarkerCluster\nimport librosa\nimport librosa.display","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.016150Z","iopub.execute_input":"2025-03-13T00:06:16.016565Z","iopub.status.idle":"2025-03-13T00:06:16.022977Z","shell.execute_reply.started":"2025-03-13T00:06:16.016525Z","shell.execute_reply":"2025-03-13T00:06:16.021376Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Settings\nsns.set_style(\"whitegrid\")\ncolor_pal = plt.rcParams[\"axes.prop_cycle\"].by_key()[\"color\"]\nwarnings.filterwarnings(\"ignore\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.026843Z","iopub.execute_input":"2025-03-13T00:06:16.027157Z","iopub.status.idle":"2025-03-13T00:06:16.045851Z","shell.execute_reply.started":"2025-03-13T00:06:16.027133Z","shell.execute_reply":"2025-03-13T00:06:16.044653Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## 📂 Prepare files\nLet's take an initial peek at the files we've been provided.\n\nThe description for each file is given on the competition data page: https://www.kaggle.com/competitions/birdclef-2025/data\n\nWe're mostly looking for any obvious ways we can clean up or join the data together before we start exploring, as well as just getting a feel for it.\n","metadata":{}},{"cell_type":"code","source":"# View files\n\ndef get_comp_files_and_dirs(input_dir):\n    file_list = []\n    dir_list = []\n    try:\n        for comp_dir in os.listdir(input_dir):\n            comp_path = '/'.join([input_dir, comp_dir])\n            print(f\"Competition Directory: {comp_path}\")\n            print(\"Contains:\")\n            with os.scandir(comp_path) as entries:\n                for entry in entries:\n                    if entry.is_file():\n                        print(f\"- (File) {entry.name}, Size: {entry.stat().st_size} bytes\")\n                        file_list.append(os.path.join(input_dir, comp_dir, entry))\n                    elif entry.is_dir():\n                        print(f\"- (Folder) {entry.name}\")\n                        dir_list.append(os.path.join(input_dir, comp_dir, entry))\n\n    except FileNotFoundError:\n        print(f\"The specified directory '{directory}' does not exist.\")\n    except PermissionError:\n        print(f\"Permission error accessing directory '{directory}'.\")\n    return file_list, dir_list\n    \ninput_dir = '/kaggle/input'\nfile_list, dir_list = get_comp_files_and_dirs(input_dir)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.052870Z","iopub.execute_input":"2025-03-13T00:06:16.053245Z","iopub.status.idle":"2025-03-13T00:06:16.083497Z","shell.execute_reply.started":"2025-03-13T00:06:16.053217Z","shell.execute_reply":"2025-03-13T00:06:16.081954Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"comp_dir = '/kaggle/input/birdclef-2025'\nss_df = pd.read_csv(os.path.join(comp_dir, 'sample_submission.csv'))\ntax_df = pd.read_csv(os.path.join(comp_dir, 'taxonomy.csv'))\ntrain_df = pd.read_csv(os.path.join(comp_dir, 'train.csv'))\nwith open(os.path.join(comp_dir, \"recording_location.txt\"), \"r\") as file:\n    recording_location = file.read()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.085153Z","iopub.execute_input":"2025-03-13T00:06:16.085554Z","iopub.status.idle":"2025-03-13T00:06:16.263886Z","shell.execute_reply.started":"2025-03-13T00:06:16.085514Z","shell.execute_reply":"2025-03-13T00:06:16.262722Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Recording Location\nThis just seems to be a general information string about where the test data was recorded.","metadata":{}},{"cell_type":"code","source":"print(recording_location)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.266151Z","iopub.execute_input":"2025-03-13T00:06:16.266570Z","iopub.status.idle":"2025-03-13T00:06:16.272375Z","shell.execute_reply.started":"2025-03-13T00:06:16.266541Z","shell.execute_reply":"2025-03-13T00:06:16.271237Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Sample Submission\nThis is an example file of what we need to submit. For each soundscape, we need to give a probability that each animal appears in that soundscape.","metadata":{}},{"cell_type":"code","source":"ss_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.274508Z","iopub.execute_input":"2025-03-13T00:06:16.274924Z","iopub.status.idle":"2025-03-13T00:06:16.311065Z","shell.execute_reply.started":"2025-03-13T00:06:16.274894Z","shell.execute_reply":"2025-03-13T00:06:16.309904Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Training Data\nThe primary label here is the main animal which is featured in each recording. There will be a bunch in the background too that haven't been labelled.\n\nThe latitude and longitude tell us where the recording was taken.","metadata":{}},{"cell_type":"code","source":"train_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.312452Z","iopub.execute_input":"2025-03-13T00:06:16.312804Z","iopub.status.idle":"2025-03-13T00:06:16.343194Z","shell.execute_reply.started":"2025-03-13T00:06:16.312776Z","shell.execute_reply":"2025-03-13T00:06:16.341866Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# How many rows are there?\nlen(train_df)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.344617Z","iopub.execute_input":"2025-03-13T00:06:16.345012Z","iopub.status.idle":"2025-03-13T00:06:16.367100Z","shell.execute_reply.started":"2025-03-13T00:06:16.344977Z","shell.execute_reply":"2025-03-13T00:06:16.365966Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Taxonomy Data\nThis is just some general info which we can append to our training data","metadata":{}},{"cell_type":"code","source":"tax_df.head(-1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.368086Z","iopub.execute_input":"2025-03-13T00:06:16.368500Z","iopub.status.idle":"2025-03-13T00:06:16.398071Z","shell.execute_reply.started":"2025-03-13T00:06:16.368462Z","shell.execute_reply":"2025-03-13T00:06:16.396549Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_df = train_df.merge(tax_df[['primary_label', 'inat_taxon_id', 'class_name']], on='primary_label', how='left')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.401750Z","iopub.execute_input":"2025-03-13T00:06:16.402146Z","iopub.status.idle":"2025-03-13T00:06:16.434467Z","shell.execute_reply.started":"2025-03-13T00:06:16.402117Z","shell.execute_reply":"2025-03-13T00:06:16.432907Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"---\n# 🧭 Explore the data\nNow let's take a closer look at some of these files!","metadata":{}},{"cell_type":"markdown","source":"## 📊 Plot the Distribution\nFrom looking at the distributions in the training data, we have lots of birds and not many frogs. It may be necessary to find additional frog data publicly available elsewhere.","metadata":{}},{"cell_type":"code","source":"count = train_df['common_name'].value_counts(ascending=True)\nplt.figure(figsize=(12, 30))\ncount.plot(kind='barh')\n\nplt.xlabel('Count')\nplt.ylabel('Label')\nplt.title('Count of Primary Species in Recordings')\nplt.tick_params(axis='x', which='both', bottom=True, top=True, labelbottom=True, labeltop=True)\n\nplt.tight_layout()\nplt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:16.435983Z","iopub.execute_input":"2025-03-13T00:06:16.436408Z","iopub.status.idle":"2025-03-13T00:06:19.218696Z","shell.execute_reply.started":"2025-03-13T00:06:16.436377Z","shell.execute_reply":"2025-03-13T00:06:19.217102Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## 🔊 Listen to Audio\nLet's hear some of these sounds already! Here's a little GUI to do that.\n","metadata":{}},{"cell_type":"markdown","source":"### Train Audio\nThese are audio files that have already been labelled. We'll use these to train our data","metadata":{}},{"cell_type":"code","source":"directory = \"../input/birdclef-2025/train_audio\"\n\nfolders_list = [f for f in os.listdir(directory) if os.path.isdir(os.path.join(directory, f))]\nprimary_label_to_name = dict(zip(tax_df['primary_label'], tax_df['common_name']))\nname_to_primary_label = dict(zip(tax_df['common_name'], tax_df['primary_label']))\n\ndef audio_gui():\n    # Create widgets\n    animal_select = widgets.Select(\n        options=sorted([primary_label_to_name.get(item, item) for item in folders_list]),\n        description='Animals:',\n        rows=10\n    )\n    \n    audio_select = widgets.Select(\n        options=[],\n        description='Audio:',\n        rows=10\n    )\n\n    audio_output = widgets.Output()\n\n    # Function to update audio list when an animal is selected\n    def on_animal_change(change):\n        audio_output.clear_output()\n        selected_animal = animal_select.value\n        if selected_animal:\n            folder_name = name_to_primary_label[selected_animal]\n            audio_list = [f for f in os.listdir(os.path.join(directory, folder_name)) if f.endswith('.ogg')]\n            audio_select.options = sorted(audio_list)  # Update the dropdown options\n\n    # Function to play selected audio\n    def on_audio_change(change):\n        audio_output.clear_output()\n        selected_animal = animal_select.value\n        selected_audio = audio_select.value\n        if selected_animal and selected_audio:\n            audio_path = os.path.join(directory, name_to_primary_label[selected_animal], selected_audio)\n            with audio_output:\n                display(IPython.display.Audio(audio_path))\n                y, sr = librosa.load(audio_path, sr=None)  # Load with original sample rate\n                \n                # Compute the spectrogram (Short-Time Fourier Transform)\n                D = librosa.stft(y)\n                DB = librosa.amplitude_to_db(np.abs(D), ref=np.max)  # Convert amplitude to dB\n                \n                # Plot the spectrogram\n                plt.figure(figsize=(20, 8))\n                librosa.display.specshow(DB, sr=sr, x_axis='time', y_axis='log')\n                plt.colorbar(label='dB')\n                plt.title(f'Spectrogram of {animal_select.value} - {audio_select.value}')\n                plt.xlabel('Time (s)')\n                plt.ylabel('Frequency (Hz)')\n                plt.show()\n\n    # Attach event listeners\n    animal_select.observe(on_animal_change, names='value')\n    audio_select.observe(on_audio_change, names='value')\n\n    # Display widgets\n    display(animal_select, audio_select, audio_output)\n\n# Run the function\naudio_gui()\n\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:38.540787Z","iopub.execute_input":"2025-03-13T00:06:38.541230Z","iopub.status.idle":"2025-03-13T00:06:38.574762Z","shell.execute_reply.started":"2025-03-13T00:06:38.541198Z","shell.execute_reply":"2025-03-13T00:06:38.573558Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Train Soundscapes\nThese are unlabelled soundscapes that are similar to what the test data will be on. It will be good to try to label these in the competition.","metadata":{}},{"cell_type":"code","source":"directory = \"../input/birdclef-2025/train_soundscapes\"\n\nfiles_list = [f for f in os.listdir(directory) if f.endswith('.ogg')]\n\ndef soundscape_gui():\n    # Create widgets\n    soundscape_select = widgets.Select(\n        options=sorted(files_list),\n        description='Soundscape:',\n        rows=10,\n        style={'description_width': 'auto'}\n    )\n    soundscape_output = widgets.Output()\n\n    # Function to play selected audio\n    def on_soundscape_change(change):\n        soundscape_output.clear_output()\n        if soundscape_select:\n            soundscape_audio_path = os.path.join(directory, soundscape_select.value)\n            with soundscape_output:\n                display(IPython.display.Audio(soundscape_audio_path))\n                y, sr = librosa.load(soundscape_audio_path, sr=None)  # Load with original sample rate\n\n                # Compute the spectrogram (Short-Time Fourier Transform)\n                D = librosa.stft(y)\n                DB = librosa.amplitude_to_db(np.abs(D), ref=np.max)  # Convert amplitude to dB\n                \n                # Plot the spectrogram\n                plt.figure(figsize=(20, 8))\n                librosa.display.specshow(DB, sr=sr, x_axis='time', y_axis='log')\n                plt.colorbar(label='dB')\n                plt.title(f'Spectrogram of {soundscape_select.value}')\n                plt.xlabel('Time (s)')\n                plt.ylabel('Frequency (Hz)')\n                plt.show()\n\n    # Attach event listeners\n    soundscape_select.observe(on_soundscape_change, names='value')\n\n    # Display widgets\n    display(soundscape_select, soundscape_output)\n    soundscape_select.layout = widgets.Layout(width='300px')\n\n# Run the function\nsoundscape_gui()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:07:50.514693Z","iopub.execute_input":"2025-03-13T00:07:50.515047Z","iopub.status.idle":"2025-03-13T00:07:50.602472Z","shell.execute_reply.started":"2025-03-13T00:07:50.515021Z","shell.execute_reply":"2025-03-13T00:07:50.601512Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## 🗺️ Recording Locations\nWe can see that the organisers have chosen audio samples mostly clustered near Colombia, but have included a smattering of other audio recordings from around the world also.\n\nFrom the looks of it, all of the animals in the database are ones that are possible to have in Colombia, so there's no obvious traps there.","metadata":{}},{"cell_type":"code","source":"data = train_df.dropna(subset=['latitude', 'longitude'])\n# Create a map centered around the mean of the data\nm = folium.Map(location=[data['latitude'].mean(), data['longitude'].mean()], zoom_start=3)\n\n# Convert DataFrame to a list of (lat, lon) tuples\nlocations = list(zip(data.latitude, data.longitude))\n\n# Add FastMarkerCluster to the map\nFastMarkerCluster(locations).add_to(m)\n\n# Display the map (work\nm","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:06:19.444207Z","iopub.execute_input":"2025-03-13T00:06:19.444596Z","iopub.status.idle":"2025-03-13T00:06:19.662468Z","shell.execute_reply.started":"2025-03-13T00:06:19.444551Z","shell.execute_reply":"2025-03-13T00:06:19.661295Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# View per animal\ndata = train_df.dropna(subset=['latitude', 'longitude'])\ndef animal_map_gui():\n    animal_map_select = widgets.Select(\n        options=tax_df['common_name'].unique(),\n        description='Animals:',\n        rows=10\n    )\n    map_output = widgets.Output()\n    display(animal_map_select)\n    def on_change(change):\n        map_output.clear_output()\n        with map_output:\n            animal_id = tax_df[tax_df['common_name']==animal_map_select.value]['primary_label'].iloc[0]\n            print(f'Name: {animal_map_select.value}')\n            print(f'ID: {animal_id}')\n            filtered_data = data[data['primary_label']==animal_id]\n            print(f'Total Recordings: {len(filtered_data)}')\n            m = folium.Map(location=[filtered_data['latitude'].mean(), filtered_data['longitude'].mean()], zoom_start=3)\n            \n            # Convert DataFrame to a list of (lat, lon) tuples\n            locations = list(zip(filtered_data.latitude, filtered_data.longitude))\n            FastMarkerCluster(locations).add_to(m)\n            display(m)\n    animal_map_select.observe(on_change, names='value')\n    display(map_output)\n            \nanimal_map_gui()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-13T00:08:06.501996Z","iopub.execute_input":"2025-03-13T00:08:06.502442Z","iopub.status.idle":"2025-03-13T00:08:06.529994Z","shell.execute_reply.started":"2025-03-13T00:08:06.502394Z","shell.execute_reply":"2025-03-13T00:08:06.528793Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"---\n# 🐦 Conclusion\nI think this will be quite an interesting competition. Because we have no idea what the test data will be like, we'll be forced to create very general models that are capable of processing unknown and potentially unclean data.\n\nIf you enjoyed this notebook, please give it an upvote and leave a comment so that other people can find it too!\n\nBest of luck Kaggling!","metadata":{}}]}