{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\n# for dirname, _, filenames in os.walk('/kaggle/input'):\n#     for filename in filenames:\n#         print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import torch\nimport torchaudio\nimport torch.nn as nn\nimport torch.nn.functional as F\nfrom torchvision.transforms import Compose, Resize\nfrom torchaudio.transforms import AmplitudeToDB, MelSpectrogram\n\nimport sys\nimport glob\n\npackage_path = \"../input/efficientnet-pytorch/EfficientNet-PyTorch/EfficientNet-PyTorch-master/\"\nsys.path.append(package_path)\nimport efficientnet_pytorch","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_audios = glob.glob(\"/kaggle/input/birdclef-2023/test_soundscapes/*\")\ntest_audios","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"target_names = [_.split(\"/\")[-1] for _ in glob.glob(\"/kaggle/input/birdclef-2023/train_audio/*\")]\ntarget_names = sorted(target_names)\nlen(target_names)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"models_path = glob.glob(\"/kaggle/input/birdclef23-trained-models/*\")\nmodels_path = sorted(models_path)\nmodels_path","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"mel_spectogram = MelSpectrogram(\n    sample_rate=32000,\n    n_fft=1024,\n    hop_length=None,\n    n_mels=64\n)\n\nimg_transform = Compose([\n    Resize((128, 384)),\n])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class Model(nn.Module):\n    def __init__(self, out_features):\n        super().__init__()\n        self.net = efficientnet_pytorch.EfficientNet.from_name(\"efficientnet-b1\")\n        n_features = self.net._fc.in_features\n        self.net._fc = nn.Linear(in_features=n_features, out_features=out_features, bias=True)\n    \n    def forward(self, x):\n        out = self.net(x)\n        return out","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"weight_path = \"/kaggle/input/birdclef23-trained-models/model6_b1.pth\"\nprint(weight_path)\nout_features = len(target_names)\n\nmodel = Model(out_features)\nmodel.load_state_dict(torch.load(weight_path, map_location=torch.device('cpu')))\nmodel.eval();","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ans = []\n\nfor audio in test_audios:\n    waveform, sr = torchaudio.load(audio)\n    time = int(waveform.shape[1]/sr)\n\n    for i in range(0, time, 5):\n        row_id = str(audio.split(\"/\")[-1].split(\".\")[0]) + \"_\" + str(i+5)\n\n        wvr = waveform[:, i*sr:(i+5)*sr]\n        spectrogram = mel_spectogram(wvr)\n        spectrogram = AmplitudeToDB(top_db=80)(spectrogram)\n        spectrogram = spectrogram.repeat(3, 1, 1)\n        spectrogram = img_transform(spectrogram)\n        spectrogram = (spectrogram - spectrogram.min()) / (spectrogram.max() - spectrogram.min())\n        pred = model(spectrogram.unsqueeze(0))\n        pred = F.softmax(pred, dim=1)[0]\n        ans.append([row_id, *pred.detach().numpy()])\n\nans = np.array(ans)\nans.shape","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission_df = pd.DataFrame(columns=[\"row_id\", *target_names])\n\nfor i in range(len(submission_df.columns)):\n    submission_df[submission_df.columns[i]] = ans[:, i] if i==0 else ans[:, i].astype(float)\n\nsubmission_df.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission_df.to_csv(\"submission.csv\", index=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}