\n \n**Objective**: The competition aims to advance biodiversity monitoring efforts in the Western Ghats, a Global Biodiversity Hotspot in India, by leveraging machine learning techniques to identify under-studied Indian bird species through their calls.\n\n**Dataset**: Participants are provided with a dataset consisting of audio recordings of bird calls from the Western Ghats region, along with associated metadata such as species labels, geographic coordinates, and recording details. Additionally, there are test soundscapes and unlabeled audio data for evaluation.\n\n**Task**: Participants are tasked with developing computational solutions to process the audio data and accurately classify bird species present in the recordings. The competition focuses on identifying endemic species, detecting endangered species with limited training data, and classifying poorly understood nocturnal species.\n\n**Collaborators**: The competition is collaboratively organized by various institutions including the Chemnitz University of Technology, Google Research, Indian Institute of Science Education and Research (IISER) Tirupati, K. Lisa Yang Center for Conservation Bioacoustics at the Cornell Lab of Ornithology, LifeCLEF, and Xeno-canto.\n\n**Significance**: Successful solutions developed in the competition will contribute to more comprehensive and efficient monitoring of avian populations in the Western Ghats. This, in turn, will aid conservation efforts by providing insights into biodiversity trends and informing targeted conservation actions.","metadata":{}},{"cell_type":"markdown","source":"
\n\n
🔱 Evalaution Metric💢 :
\n \n\n","metadata":{},"attachments":{"4b0301f9-5443-4215-b513-040d227ae166.png":{"image/png":"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"}}},{"cell_type":"code","source":"import sys, os\nsys.path.append('/kaggle/input/efficientnet-keras-dataset/efficientnet_kaggle')\n!pip install -q /kaggle/input/tensorflow-extra-lib-ds/tensorflow_extra-1.0.2-py3-none-any.whl --no-deps","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:44:30.981900Z","iopub.execute_input":"2024-07-01T16:44:30.982337Z","iopub.status.idle":"2024-07-01T16:44:54.046570Z","shell.execute_reply.started":"2024-07-01T16:44:30.982304Z","shell.execute_reply":"2024-07-01T16:44:54.045044Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import tensorflow as tf\ntf.get_logger().setLevel('ERROR')\ntf.autograph.set_verbosity(0)\nimport os\nimport pandas as pd\nimport numpy as np\nimport random\nfrom glob import glob\nfrom tqdm import tqdm\ntqdm.pandas()\nimport gc\nimport librosa\nimport sklearn\nimport time\nimport glob\nimport shutil\n\nimport tensorflow as tf\ntf.config.optimizer.set_jit(True) \nimport tensorflow_io as tfio\nimport tensorflow.keras.backend as K\nimport matplotlib.pyplot as plt\n\nimport efficientnet.tfkeras as efn\nimport tensorflow_extra as tfe\nimport warnings\nwarnings.filterwarnings('ignore')\nimport seaborn as sns\nfrom matplotlib.offsetbox import OffsetImage, AnnotationBbox\nimport matplotlib.image as mpimg\n","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:44:54.049520Z","iopub.execute_input":"2024-07-01T16:44:54.050042Z","iopub.status.idle":"2024-07-01T16:45:15.148409Z","shell.execute_reply.started":"2024-07-01T16:44:54.049995Z","shell.execute_reply":"2024-07-01T16:45:15.146630Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n
💡 About Bird Species :
\n \n**Diversity**: India boasts over 1,300 species of birds, reflecting its rich avifauna.\n\n**Endemism**: The country is home to several endemic bird species found nowhere else, such as the Indian Peafowl and the Nilgiri Laughingthrush.\n\n**Migration**: India serves as a vital wintering ground for migratory birds, attracting species from various regions during the winter months.\n\n**Threats**: Habitat loss, poaching, pollution, and climate change pose significant threats to bird species in India, with many being classified as endangered or vulnerable.\n\n**Cultural Importance**: Birds hold cultural and religious significance in Indian mythology, folklore, and art, with iconic species like the Indian Peafowl being national symbols.\n\n**Conservation**: Conservation efforts in India include the establishment of protected areas, wildlife sanctuaries, and conservation projects aimed at safeguarding bird species and their habitats\n","metadata":{}},{"cell_type":"code","source":"df = pd.read_csv('/kaggle/input/birdclef-2024/train_metadata.csv')","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:15.150358Z","iopub.execute_input":"2024-07-01T16:45:15.151126Z","iopub.status.idle":"2024-07-01T16:45:15.381475Z","shell.execute_reply.started":"2024-07-01T16:45:15.151084Z","shell.execute_reply":"2024-07-01T16:45:15.379834Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df.head()","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:15.385147Z","iopub.execute_input":"2024-07-01T16:45:15.385744Z","iopub.status.idle":"2024-07-01T16:45:15.428670Z","shell.execute_reply.started":"2024-07-01T16:45:15.385668Z","shell.execute_reply":"2024-07-01T16:45:15.427370Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df.columns","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:15.430417Z","iopub.execute_input":"2024-07-01T16:45:15.431743Z","iopub.status.idle":"2024-07-01T16:45:15.445106Z","shell.execute_reply.started":"2024-07-01T16:45:15.431692Z","shell.execute_reply":"2024-07-01T16:45:15.443825Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df.dtypes","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:15.447457Z","iopub.execute_input":"2024-07-01T16:45:15.449179Z","iopub.status.idle":"2024-07-01T16:45:15.464268Z","shell.execute_reply.started":"2024-07-01T16:45:15.449136Z","shell.execute_reply":"2024-07-01T16:45:15.462786Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"bird_path = '/kaggle/input/birdcall-recognition-data/blue bird.jpg'\nbird = mpimg.imread(bird_path)\nimagebox = OffsetImage(bird, zoom=0.5)\n\nxy = (0.5, 0.7)\nab = AnnotationBbox(imagebox, xy, frameon=False, pad=1, xybox=(130, 300))\nlabel_counts = df['primary_label'].value_counts()\nplt.figure(figsize=(45, 10))\nax = sns.barplot(x=label_counts.index, y=label_counts.values, palette=\"hls\")\nax.add_artist(ab)\n\n\nplt.title(\"Bird Species\", fontsize=16)\nplt.xticks(rotation=90, fontsize=15)\nplt.yticks(fontsize=13)\nplt.ylabel(\"Frequency\", fontsize=20)\nplt.xlabel(\"Bird Species\", fontsize=20)\nplt.gcf().set_facecolor('lightgrey') # Set background color of the entire figure\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:15.466306Z","iopub.execute_input":"2024-07-01T16:45:15.467020Z","iopub.status.idle":"2024-07-01T16:45:18.161970Z","shell.execute_reply.started":"2024-07-01T16:45:15.466952Z","shell.execute_reply":"2024-07-01T16:45:18.160382Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"bird_path = '/kaggle/input/birdcall-recognition-data/fluff ball.jpg'\nbird = mpimg.imread(bird_path)\nimagebox = OffsetImage(bird, zoom=0.5)\n\nxy = (0.5, 0.7)\nab = AnnotationBbox(imagebox, xy, frameon=False, pad=1, xybox=(13, 100))\nplt.figure(figsize=(24, 16))\nsns.scatterplot(x='latitude', y='longitude', data=df, color='skyblue')\nplt.gca().add_artist(ab)\n\nplt.title(\"Geographic Distribution of Recordings\", fontsize=30)\nplt.xlabel(\"Latitude\", fontsize=30)\nplt.ylabel(\"Longitude\", fontsize=30)\nplt.xticks(fontsize=20)\nplt.yticks(fontsize=20)\nplt.gcf().set_facecolor('lightgrey') # Set background color of the entire figure\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:18.163566Z","iopub.execute_input":"2024-07-01T16:45:18.164028Z","iopub.status.idle":"2024-07-01T16:45:18.984404Z","shell.execute_reply.started":"2024-07-01T16:45:18.163970Z","shell.execute_reply":"2024-07-01T16:45:18.983211Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"bird_path = '/kaggle/input/birdcall-recognition-data/green bird.jpg'\nbird = mpimg.imread(bird_path)\nimagebox = OffsetImage(bird, zoom=0.5)\n\nxy = (0.5, 0.7)\nab = AnnotationBbox(imagebox, xy, frameon=False, pad=1, xybox=(3,15000))\nplt.figure(figsize=(30,15))\nsns.countplot(x='license', data=df, palette='hls')\nplt.gca().add_artist(ab)\nplt.title(\"Distribution of Recording Licenses\", fontsize=20)\nplt.xlabel(\"License\", fontsize=20)\nplt.ylabel(\"Frequency\", fontsize=20)\nplt.xticks(rotation=45, fontsize=20)\nplt.yticks(fontsize=20)\nplt.gcf().set_facecolor('lightgrey') # Set background color of the entire figure\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:18.985673Z","iopub.execute_input":"2024-07-01T16:45:18.986053Z","iopub.status.idle":"2024-07-01T16:45:19.775043Z","shell.execute_reply.started":"2024-07-01T16:45:18.986023Z","shell.execute_reply":"2024-07-01T16:45:19.773786Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import random\nunique_authors = df['author'].unique()\nselected_authors = random.sample(list(unique_authors), k=40) \ndf_selected_authors = df[df['author'].isin(selected_authors)]\n\n\nbird_path = '/kaggle/input/birdcall-recognition-data/multicolor bird.jpg'\nbird = mpimg.imread(bird_path)\nimagebox = OffsetImage(bird, zoom=0.5)\n\nxy = (0.5, 0.7)\nab = AnnotationBbox(imagebox, xy, frameon=False, pad=1, xybox=(30,45))\n\n\nplt.figure(figsize=(45, 15))\nsns.countplot(x='author', data=df_selected_authors, palette='hls')\nplt.gca().add_artist(ab)\nplt.title(\"Distribution of Recording Authors\", fontsize=30)\nplt.xlabel(\"Author\", fontsize=30)\nplt.ylabel(\"Frequency\", fontsize=30)\nplt.xticks(rotation=90, fontsize=30)\nplt.yticks(fontsize=20)\nplt.gcf().set_facecolor('lightgrey') # Set background color of the entire figure\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:19.778812Z","iopub.execute_input":"2024-07-01T16:45:19.779210Z","iopub.status.idle":"2024-07-01T16:45:21.232293Z","shell.execute_reply.started":"2024-07-01T16:45:19.779178Z","shell.execute_reply":"2024-07-01T16:45:21.230936Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"","metadata":{},"attachments":{"8f75f9b7-4193-45a2-a5e1-27f5e6a2d634.jpg":{"image/jpeg":"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"}}},{"cell_type":"markdown","source":"Malabar Grey Hornbill (Ocyceros griseus):\nProtection of habitat through preserving old-growth forests.\nCombating deforestation, habitat fragmentation, and illegal hunting.\nNilgiri Wood Pigeon (Columba elphinstonii):\nProtecting montane forest habitat from deforestation and degradation.\nFocusing on preserving connectivity between fragmented forest patches.\nBlue-winged Parakeet (Psittacula columboides):\nProtecting forest habitat and nesting sites.\nAddressing threats such as deforestation, habitat loss, and illegal trade.\nNilgiri Pipit (Anthus nilghiriensis):\nPreservation and restoration of montane grassland habitats.\nSustainable land management practices and protection from invasive species.\nBlack-and-orange Flycatcher (Ficedula nigrorufa):\nProtecting forest habitat from deforestation and fragmentation.\nMaintaining forest integrity and mitigating climate change impacts.\nNilgiri Flycatcher (Eumyias albicaudatus):\nPreserving shola forest habitat and montane grasslands.\nProtecting habitats from degradation, invasive species, and human disturbances.\nWhite-bellied Blue Robin (Sholicola albiventris):\nProtection of montane forest habitat from deforestation and fragmentation.\nPreserving the integrity of shola forests and mitigating human impacts.\nRufous Babbler (Turdoides subrufa):\nPreservation of forest habitat and protection from degradation.\nMaintaining connectivity between fragmented forest patches and addressing threats like deforestation and invasive species.\nGrey-headed Bulbul (Pycnonotus priocephalus):\nProtecting forest habitat from deforestation and habitat degradation.\nPreserving the integrity of evergreen forests and mitigating human impacts.\nYellow-browed Bulbul (Acritillas indica):\nProtection of forest habitat from deforestation and fragmentation.\nPreserving biodiversity-rich habitats and mitigating human disturbances.\nWhite-cheeked Barbet (Psilopogon viridis):\nProtecting forest habitat from deforestation and habitat degradation.\nPreserving biodiversity-rich habitats and mitigating human impacts.","metadata":{}},{"cell_type":"code","source":"import random\nnum_ratings = random.randint(2, 5) \nselected_ratings = random.sample(list(df['rating'].unique()), k=num_ratings)\ndf_selected_ratings = df[df['rating'].isin(selected_ratings)]\npalette = sns.color_palette(\"hls\", n_colors=num_ratings)\n\n\nbird_path = '/kaggle/input/birdcall-recognition-data/violet bird.jpg'\nbird = mpimg.imread(bird_path)\nimagebox = OffsetImage(bird, zoom=0.5)\nxy = (0.5, 0.7)\nab = AnnotationBbox(imagebox, xy, frameon=False, pad=1, xybox=(0.5,800))\n\n\nplt.figure(figsize=(45, 15))\nsns.countplot(x='rating', data=df_selected_ratings, palette=palette)\nplt.gca().add_artist(ab)\n\n\nplt.title(\"Distribution of Ratings\", fontsize=20)\nplt.xlabel(\"Rating\", fontsize=20)\nplt.ylabel(\"Frequency\", fontsize=20)\nplt.xticks(rotation=45, fontsize=20)\nplt.yticks(fontsize=20)\nplt.gcf().set_facecolor('#CCCCFF') # Set background color of the entire figure\n\nplt.show()\n","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:21.233819Z","iopub.execute_input":"2024-07-01T16:45:21.234261Z","iopub.status.idle":"2024-07-01T16:45:21.947134Z","shell.execute_reply.started":"2024-07-01T16:45:21.234225Z","shell.execute_reply":"2024-07-01T16:45:21.945698Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#
2 | EXPLORATORY DATA ANALYSIS\n
| Configuration
\n
\n
Parames Sets on Dataset
","metadata":{}},{"cell_type":"code","source":"class CFG:\n debug = False\n verbose = 0\n \n device = 'CPU'\n seed = 42\n \n \n img_size = [128, 384]\n batch_size = 16\n infer_bs = 2\n tta = 1\n drop_remainder = True\n \n \n duration = 5 \n train_duration = 10\n sample_rate = 32000\n downsample = 1\n audio_len = duration*sample_rate\n nfft = 2028\n window = 2048\n hop_length = train_duration*32000 // (img_size[1] - 1)\n fmin = 20\n fmax = 16000\n normalize = True\n\n \n class_names = sorted(os.listdir('/kaggle/input/birdclef-2024/train_audio/'))\n num_classes = len(class_names)\n class_labels = list(range(num_classes))\n label2name = dict(zip(class_labels, class_names))\n name2label = {v:k for k,v in label2name.items()}\n \n target_col = ['target']\n tab_cols = ['filename','common_name','rate']","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:21.949275Z","iopub.execute_input":"2024-07-01T16:45:21.949762Z","iopub.status.idle":"2024-07-01T16:45:21.982610Z","shell.execute_reply.started":"2024-07-01T16:45:21.949722Z","shell.execute_reply":"2024-07-01T16:45:21.981541Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def load_audio(filepath, sr=32000, normalize=True):\n audio, orig_sr = librosa.load(filepath, sr=None)\n if sr!=orig_sr:\n audio = librosa.resample(y, orig_sr, sr)\n audio = audio.astype('float32').ravel()\n audio = tf.convert_to_tensor(audio)\n return audio\n\n@tf.function(jit_compile=True)\ndef MakeFrame(audio, duration=5, sr=32000):\n frame_length = int(duration * sr)\n frame_step = int(duration * sr)\n chunks = tf.signal.frame(audio, frame_length, frame_step, pad_end=True)\n return chunks","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:21.984032Z","iopub.execute_input":"2024-07-01T16:45:21.984381Z","iopub.status.idle":"2024-07-01T16:45:21.993312Z","shell.execute_reply.started":"2024-07-01T16:45:21.984351Z","shell.execute_reply":"2024-07-01T16:45:21.992148Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n
💡 About The audio and frame :
\n \n**load_audio(filepath, sr=32000, normalize=True)**: This function is responsible for loading an audio file from the given filepath using the librosa library. It takes three arguments: filepath (the path to the audio file), sr (the target sampling rate, default is 32000), and normalize (a boolean indicating whether to normalize the audio data, default is True). It returns a TensorFlow tensor containing the audio data loaded from the file. If the specified sr is different from the original sampling rate of the audio file, it resamples the audio data to match the target sampling rate.\n\n**MakeFrame(audio, duration=5, sr=32000)**: This function takes an audio tensor as input and divides it into smaller chunks or frames. It's commonly used for audio segmentation tasks. The arguments are audio (the input audio tensor), duration (the duration of each frame in seconds, default is 5 seconds), and sr (the sampling rate, default is 32000). It returns a tensor containing the audio frames, where each row represents a frame of audio data.","metadata":{}},{"cell_type":"code","source":"def generate_predictions(test_df, CKPTS, NUM_CKPTS):\n ids = []\n preds = np.empty(shape=(0, 182), dtype='float32')\n \n for filepath in tqdm(test_df.filepath.tolist(), 'test '):\n filename = filepath.split('/')[-1].replace('.ogg','')\n audio = load_audio(filepath)\n chunks = MakeFrame(audio)\n \n chunk_preds = np.zeros(shape=(len(chunks), 182), dtype=np.float32)\n for model in CKPTS[:NUM_CKPTS]:\n rec_preds = model(chunks, training=False).numpy()\n chunk_preds += rec_preds / len(CKPTS)\n \n rec_ids = [f'{filename}_{(frame_id+1)*5}' for frame_id in range(len(chunks))]\n ids += rec_ids\n preds = np.concatenate([preds, chunk_preds], axis=0)\n \n return ids, preds","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:21.994737Z","iopub.execute_input":"2024-07-01T16:45:21.995122Z","iopub.status.idle":"2024-07-01T16:45:22.007127Z","shell.execute_reply.started":"2024-07-01T16:45:21.995091Z","shell.execute_reply":"2024-07-01T16:45:22.005867Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n
💡 About The generate prediction :
\n \nThis function generate_predictions takes a DataFrame containing file paths (test_df), a list of trained models (CKPTS), and the number of checkpoints to use (NUM_CKPTS). It iterates over each audio file in test_df, loads the audio, segments it into smaller frames, predicts labels for each frame using the provided models, and aggregates the predictions. Finally, it returns two lists: ids containing unique identifiers for each frame and preds containing the corresponding predicted probabilities for each class.","metadata":{}},{"cell_type":"code","source":"CKPT_DIR = '/kaggle/input/birdclef24-pretraining-train-model'\nCKPT_PATHS = sorted([x for x in glob.glob(f'{CKPT_DIR}/fold-*keras')])\nprint(\"Checkpoints: \", CKPT_PATHS)\n\n\nWRITABLE_DIR = '/kaggle/working/'\nif not os.path.exists(WRITABLE_DIR):\n os.makedirs(WRITABLE_DIR)\n\nfor ckpt_path in CKPT_PATHS:\n shutil.copy(ckpt_path, WRITABLE_DIR)\n\nCKPT_PATHS = sorted([f'{WRITABLE_DIR}/{os.path.basename(x)}' for x in glob.glob(f'{CKPT_DIR}/fold-*keras')])\nCKPTS = [tf.keras.models.load_model(x, compile=False) for x in tqdm(CKPT_PATHS, desc=\"Loading ckpts \")]\nNUM_CKPTS = 1\n\ntest_audio_dir = '/kaggle/input/birdclef-2024/test_soundscapes/'\ntest_paths = [test_audio_dir+f for f in sorted(os.listdir(test_audio_dir))]\nif len(test_paths) == 1:\n test_audio_dir = '/kaggle/input/birdclef-2024/unlabeled_soundscapes/'\n test_paths = [test_audio_dir+f for f in sorted(os.listdir(test_audio_dir))][:2]\n\ntest_df = pd.DataFrame(test_paths, columns=['filepath'])\ntest_df['filename'] = test_df.filepath.map(lambda x: x.split('/')[-1].replace('.ogg',''))\n\n#ids, preds = generate_predictions(test_df, CKPTS, NUM_CKPTS)","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:22.008421Z","iopub.execute_input":"2024-07-01T16:45:22.008778Z","iopub.status.idle":"2024-07-01T16:45:30.513015Z","shell.execute_reply.started":"2024-07-01T16:45:22.008749Z","shell.execute_reply":"2024-07-01T16:45:30.511672Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def predict_and_ensemble(test_df, load_audio, MakeFrame, CKPTS, NUM_CKPTS):\n tick = time.time()\n ids = []\n preds1 = np.empty(shape=(0, 182), dtype='float32')\n\n for filepath in tqdm(test_df.filepath.tolist(), 'test '):\n filename = filepath.split('/')[-1].replace('.ogg','')\n audio = load_audio(filepath)\n chunks = MakeFrame(audio)\n\n chunk_preds = np.zeros(shape=(len(chunks), 182), dtype=np.float32)\n for model in CKPTS[:NUM_CKPTS]:\n rec_preds = model(chunks, training=False).numpy()\n chunk_preds += rec_preds / len(CKPTS)\n\n rec_ids = [f'{filename}_{(frame_id+1)*5}' for frame_id in range(len(chunks))]\n ids += rec_ids\n preds1 = np.concatenate([preds1, chunk_preds], axis=0)\n\n tock = time.time()\n execution_time = tock - tick\n sub_time = time.gmtime(execution_time * 550) \n sub_time = time.strftime(\"%H hr: %M min : %S sec\", sub_time)\n print(f\">> Time for submission: ~ {sub_time}\")\n\n return ids, preds1, execution_time\n","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:30.514551Z","iopub.execute_input":"2024-07-01T16:45:30.515039Z","iopub.status.idle":"2024-07-01T16:45:30.527409Z","shell.execute_reply.started":"2024-07-01T16:45:30.514971Z","shell.execute_reply":"2024-07-01T16:45:30.525955Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ids, preds1, execution_time = predict_and_ensemble(test_df, load_audio, MakeFrame, CKPTS, NUM_CKPTS)\n\nprint(\"IDs:\", ids)\nprint(\"Predictions:\", preds1)\nprint(\"Execution Time:\", execution_time)","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:30.529161Z","iopub.execute_input":"2024-07-01T16:45:30.529570Z","iopub.status.idle":"2024-07-01T16:45:52.901894Z","shell.execute_reply.started":"2024-07-01T16:45:30.529537Z","shell.execute_reply":"2024-07-01T16:45:52.900712Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#
1 | DATA CLEANSING AND DATA PREP\n
Final Submission
","metadata":{}},{"cell_type":"code","source":"pred_df = pd.DataFrame(ids, columns=['row_id'])\npred_df.loc[:, CFG.class_names] = preds1\npred_df.to_csv('submission.csv',index=False)\npred_df","metadata":{"execution":{"iopub.status.busy":"2024-07-01T16:45:52.903334Z","iopub.execute_input":"2024-07-01T16:45:52.904039Z","iopub.status.idle":"2024-07-01T16:45:53.090778Z","shell.execute_reply.started":"2024-07-01T16:45:52.904006Z","shell.execute_reply":"2024-07-01T16:45:53.089514Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Thank You! 🙏\n\n
\n ༼ つ ◕‿◕ ༽つ Thank You!\n
\n\n
\n 💌 Thanks for your time and attention! If you have any feedback or suggestions, please share them.
\n 🚀 If you enjoyed this notebook, consider giving it an upvote to help others find it too. Your support means a lot!
\n ❤️ Once again, thank you for your support, and I look forward to seeing you again soon!\n
","metadata":{"execution":{"iopub.status.busy":"2024-04-09T10:54:43.663921Z","iopub.execute_input":"2024-04-09T10:54:43.664340Z","iopub.status.idle":"2024-04-09T10:54:43.672753Z","shell.execute_reply.started":"2024-04-09T10:54:43.664302Z","shell.execute_reply":"2024-04-09T10:54:43.671658Z"}}},{"cell_type":"markdown","source":"
\n##
Refrences
","metadata":{}},{"cell_type":"markdown","source":"
⚡🏍Thanks to Aikhmelnyt for work i have done little restructing of code .https://www.kaggle.com/code/aikhmelnytskyy/birdclef24-pretraining-is-all-you-need-infer
","metadata":{}}]}