{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Resnet 34"},{"metadata":{"trusted":true},"cell_type":"code","source":"import pandas as pd\nimport os\nimport matplotlib.pyplot as plt\n# ../input/test-data-1-spec-rainforest/datasets/TrainingSet/spec/a5eaa2ad5/0024_037.png\nTEST_PATH='../input/test-data-1-spec-rainforest/datasets/TrainingSet/spec'\nsub_csv=pd.read_csv('../input/rfcx-species-audio-detection/sample_submission.csv')\nsub_csv_copy=sub_csv.copy()\ndf_test_data=pd.read_csv('../input/test-data-1-spec-rainforest/test_data_1_list.csv')\nsub_csv","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_test_data","execution_count":null,"outputs":[]},{"metadata":{"_kg_hide-output":true,"trusted":true},"cell_type":"code","source":"!pip3 install keras-resnet\n!pip install np_utils","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.models import load_model\nfrom keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n\nimport keras\nfrom keras.utils.np_utils import to_categorical\nimport keras_resnet.models\n\nshape, classes = (128, 256, 1), 24\n\nx = keras.layers.Input(shape)\n\nmodel = keras_resnet.models.ResNet34(x, classes=classes)\n\nmodel.compile(\"adam\", \"categorical_crossentropy\", [\"accuracy\"])\n#model = load_model('../input/train-rainforest/weights-improvement-18-0.80.hdf5')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.models import Sequential\nfrom keras.layers import Dense\nfrom keras.models import model_from_json\nimport numpy as np\n# load json and create model\n# json_file = open('../input/rainforest-train-py/model.json', 'r')\n# loaded_model_json = json_file.read()\n# json_file.close()\n# loaded_model = model_from_json(loaded_model_json)\n# load weights into new model\nmodel.load_weights(\"../input/train-rainforest-resnet34/weights-improvement-20-0.79.hdf5\")\nprint(\"Loaded model from disk\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def getRandomState():\n    RANDOM_SEED = 1\n    RANDOM = np.random.RandomState(RANDOM_SEED)\n    return RANDOM","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def prepare_data(path):\n    x_train=plt.imread(path)\n    x_train = x_train.reshape(128, 256, 1) \n    return x_train","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"count=0\nfor audio_id in sub_csv['recording_id']:\n    print(count)\n    count+=1\n    AUDIO_CLASS = os.path.join(TEST_PATH, audio_id)\n    #print(AUDIO_CLASS)\n    x_onefile_train=[]\n    all_audio=os.listdir(AUDIO_CLASS)  \n    for one_sec_audio in all_audio:\n        one_sec_audio_path=os.path.join(AUDIO_CLASS,one_sec_audio)\n        x_train = prepare_data(one_sec_audio_path)\n        x_onefile_train.append(x_train)\n\n    x_onefile_train=np.array(x_onefile_train)\n    predictions = model.predict(x_onefile_train)\n    #print(predictions.shape)\n    prob=np.max(predictions, axis=0)\n    prob=prob.reshape(1,24)\n    #print(prob)\n    \n    sub = pd.DataFrame(prob, columns=[f\"s{i}\" for i in range(24)])\n    sub[\"recording_id\"] = audio_id\n    sub = sub[[\"recording_id\"] + [f\"s{i}\" for i in range(24)]]\n    sub_csv_copy.set_index('recording_id', inplace=True)\n    sub_csv_copy.update(sub.set_index('recording_id'))\n    sub_csv_copy=sub_csv_copy.reset_index()\n#     print(sub)\n#     print(audio_id)\n#     print(prob)\n#     print(\"---------------------------------------------------------------\")\n    \n    \n        #print(one_sec_audio_path)\n            ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sub","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sub_csv_copy","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sub_csv_copy.to_csv(\"submission.csv\", index=False)","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}