{"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":"markdown","source":"

📝 Abstract

\n\n * In this Inference Notebook EfficientNetB1 is inferred.\n\n**Versions** \n* Version 3: EfficientNetB1 512 Augs CV:- **0.8684** LB:- **0.8716**\n\n\n\n\n\n**References are mentioned at the end.**\n","metadata":{}},{"cell_type":"markdown","source":"\n

📄 Table of Contents

\n\n* [Imports](#1)\n* [Configs](#2)\n* [Reading TFRecords](#3) \n* [Dataset Creation](#4)\n* [Modelling](#5)\n* [KFold Prediction](#6)\n* [References](#7)","metadata":{}},{"cell_type":"markdown","source":" \n

🚚 Imports

","metadata":{}},{"cell_type":"code","source":"!pip install efficientnet tensorflow_addons > /dev/null","metadata":{"id":"8SuhhfuItC_p","execution":{"iopub.status.busy":"2021-09-17T18:44:23.174617Z","iopub.execute_input":"2021-09-17T18:44:23.175255Z","iopub.status.idle":"2021-09-17T18:44:34.221262Z","shell.execute_reply.started":"2021-09-17T18:44:23.175155Z","shell.execute_reply":"2021-09-17T18:44:34.220365Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os \nimport re\nimport warnings\nfrom glob import glob\nfrom tqdm.auto import tqdm \nfrom scipy.signal import get_window\nfrom typing import Optional, Tuple\n\n\nimport numpy as np \nimport pandas as pd\n\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom sklearn.model_selection import train_test_split,StratifiedKFold,KFold\n\nimport tensorflow as tf\nimport tensorflow.keras.backend as K\nfrom tensorflow.keras.models import Model , load_model, save_model \nfrom tensorflow.keras.layers import Input,Dropout,Conv2D,Flatten,GlobalAveragePooling2D,Dense,GlobalAvgPool2D\nimport efficientnet.tfkeras as efn\nfrom tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, EarlyStopping\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.losses import BinaryCrossentropy\nimport tensorflow_addons as tfa\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","id":"N4JJaCzwTx1n","execution":{"iopub.status.busy":"2021-09-17T18:44:34.223377Z","iopub.execute_input":"2021-09-17T18:44:34.223623Z","iopub.status.idle":"2021-09-17T18:44:41.447712Z","shell.execute_reply.started":"2021-09-17T18:44:34.223595Z","shell.execute_reply":"2021-09-17T18:44:41.446947Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

⚙️ Configs

","metadata":{}},{"cell_type":"code","source":"# Detect hardware, return appropriate distribution strategy\ntry:\n # TPU detection. No parameters necessary if TPU_NAME environment variable is\n # set: this is always the case on Kaggle.\n tpu = tf.distribute.cluster_resolver.TPUClusterResolver()\n print('Running on TPU ', tpu.master())\nexcept ValueError:\n tpu = None\n\nif tpu:\n tf.config.experimental_connect_to_cluster(tpu)\n tf.tpu.experimental.initialize_tpu_system(tpu)\n strategy = tf.distribute.experimental.TPUStrategy(tpu)\nelse:\n # Default distribution strategy in Tensorflow. Works on CPU and single GPU.\n strategy = tf.distribute.get_strategy()\n\nprint(\"REPLICAS: \", strategy.num_replicas_in_sync)","metadata":{"id":"H6e594SfTx1u","outputId":"8cf46a05-6c00-4379-90cf-53a9c5032b82","execution":{"iopub.status.busy":"2021-09-17T18:44:41.449009Z","iopub.execute_input":"2021-09-17T18:44:41.449870Z","iopub.status.idle":"2021-09-17T18:44:47.058301Z","shell.execute_reply.started":"2021-09-17T18:44:41.449825Z","shell.execute_reply":"2021-09-17T18:44:47.057582Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class config:\n seed = 2021\n # GCS_PATH = KaggleDatasets().get_gcs_path(\"cqt-g2net-v2-0-1\")\n replicas = strategy.num_replicas_in_sync\n batch_size = 32 * replicas\n AUTO = tf.data.experimental.AUTOTUNE\n dir_path = './result'\n epochs = 20\n #augmentations\n image_size = 512\n aug = True \n mix_up_p = 0.1\n s_shift = 0.0\n t_shift = 0.0\n r_angle = 0 / 180 * np.pi\n label_positive_shift = 0.99\n\nif not os.path.exists(config.dir_path):\n os.makedirs(config.dir_path)","metadata":{"id":"1m-ElibETx1y","execution":{"iopub.status.busy":"2021-09-17T18:44:47.059996Z","iopub.execute_input":"2021-09-17T18:44:47.060733Z","iopub.status.idle":"2021-09-17T18:44:47.068827Z","shell.execute_reply.started":"2021-09-17T18:44:47.060692Z","shell.execute_reply":"2021-09-17T18:44:47.067918Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"IMAGE_SIZE = 256\nBATCH_SIZE = 32\nEFFICIENTNET_SIZE = 1\nWEIGHTS = \"imagenet\"\nAUTO = tf.data.experimental.AUTOTUNE\nREPLICAS = strategy.num_replicas_in_sync","metadata":{"id":"6K_V6llATx10","execution":{"iopub.status.busy":"2021-09-17T18:44:47.069995Z","iopub.execute_input":"2021-09-17T18:44:47.070243Z","iopub.status.idle":"2021-09-17T18:44:47.081993Z","shell.execute_reply.started":"2021-09-17T18:44:47.070209Z","shell.execute_reply":"2021-09-17T18:44:47.081260Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Dataset of Hidehisa Arai\n\ntrain_files = ['gs://kds-fe725ba5a3259c712812aed413cfe61fc3827b135988e28e694e515c/train0.tfrecords',\n 'gs://kds-fe725ba5a3259c712812aed413cfe61fc3827b135988e28e694e515c/train1.tfrecords',\n 'gs://kds-fe725ba5a3259c712812aed413cfe61fc3827b135988e28e694e515c/train2.tfrecords',\n 'gs://kds-fe725ba5a3259c712812aed413cfe61fc3827b135988e28e694e515c/train3.tfrecords',\n 'gs://kds-fe725ba5a3259c712812aed413cfe61fc3827b135988e28e694e515c/train4.tfrecords',\n 'gs://kds-ecd1c4515f350d79d5a8d2e9a07df52885278c6cce43ad116c8b03cb/train5.tfrecords',\n 'gs://kds-ecd1c4515f350d79d5a8d2e9a07df52885278c6cce43ad116c8b03cb/train6.tfrecords',\n 'gs://kds-ecd1c4515f350d79d5a8d2e9a07df52885278c6cce43ad116c8b03cb/train7.tfrecords',\n 'gs://kds-ecd1c4515f350d79d5a8d2e9a07df52885278c6cce43ad116c8b03cb/train8.tfrecords',\n 'gs://kds-ecd1c4515f350d79d5a8d2e9a07df52885278c6cce43ad116c8b03cb/train9.tfrecords',\n 'gs://kds-2c0b3b314d8607851f6dddea1e976aab52547b68de5117cf3e76a85a/train10.tfrecords',\n 'gs://kds-2c0b3b314d8607851f6dddea1e976aab52547b68de5117cf3e76a85a/train11.tfrecords',\n 'gs://kds-2c0b3b314d8607851f6dddea1e976aab52547b68de5117cf3e76a85a/train12.tfrecords',\n 'gs://kds-2c0b3b314d8607851f6dddea1e976aab52547b68de5117cf3e76a85a/train13.tfrecords',\n 'gs://kds-2c0b3b314d8607851f6dddea1e976aab52547b68de5117cf3e76a85a/train14.tfrecords',\n 'gs://kds-965e4184655c915b40b37bf7e5d3c8e310dd5751e8a2e1f5bdc8d70d/train15.tfrecords',\n 'gs://kds-965e4184655c915b40b37bf7e5d3c8e310dd5751e8a2e1f5bdc8d70d/train16.tfrecords',\n 'gs://kds-965e4184655c915b40b37bf7e5d3c8e310dd5751e8a2e1f5bdc8d70d/train17.tfrecords',\n 'gs://kds-965e4184655c915b40b37bf7e5d3c8e310dd5751e8a2e1f5bdc8d70d/train18.tfrecords',\n 'gs://kds-965e4184655c915b40b37bf7e5d3c8e310dd5751e8a2e1f5bdc8d70d/train19.tfrecords']\n\ntest_files = ['gs://kds-3af0f2c792ef9e6f5d3c87bb80c99a263fc9707106cd595aaf02e3eb/test0.tfrecords',\n 'gs://kds-3af0f2c792ef9e6f5d3c87bb80c99a263fc9707106cd595aaf02e3eb/test1.tfrecords',\n 'gs://kds-3af0f2c792ef9e6f5d3c87bb80c99a263fc9707106cd595aaf02e3eb/test2.tfrecords',\n 'gs://kds-3af0f2c792ef9e6f5d3c87bb80c99a263fc9707106cd595aaf02e3eb/test3.tfrecords',\n 'gs://kds-3af0f2c792ef9e6f5d3c87bb80c99a263fc9707106cd595aaf02e3eb/test4.tfrecords',\n 'gs://kds-9eb3135b732342c0aec8339381fec9fd19f8a7ad94ca9c31ad51fe2b/test5.tfrecords',\n 'gs://kds-9eb3135b732342c0aec8339381fec9fd19f8a7ad94ca9c31ad51fe2b/test6.tfrecords',\n 'gs://kds-9eb3135b732342c0aec8339381fec9fd19f8a7ad94ca9c31ad51fe2b/test7.tfrecords',\n 'gs://kds-9eb3135b732342c0aec8339381fec9fd19f8a7ad94ca9c31ad51fe2b/test8.tfrecords',\n 'gs://kds-9eb3135b732342c0aec8339381fec9fd19f8a7ad94ca9c31ad51fe2b/test9.tfrecords']","metadata":{"id":"1nWPTD4pUL3x","execution":{"iopub.status.busy":"2021-09-17T18:44:47.084510Z","iopub.execute_input":"2021-09-17T18:44:47.084952Z","iopub.status.idle":"2021-09-17T18:44:47.095780Z","shell.execute_reply.started":"2021-09-17T18:44:47.084909Z","shell.execute_reply":"2021-09-17T18:44:47.094700Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Creation of on fly CQT

","metadata":{}},{"cell_type":"code","source":"# Function to create cqt kernel\ndef create_cqt_kernels(\n q: float,\n fs: float,\n fmin: float,\n n_bins: int = 84,\n bins_per_octave: int = 12,\n norm: float = 1,\n window: str = \"hann\",\n fmax: Optional[float] = None,\n topbin_check: bool = True\n) -> Tuple[np.ndarray, int, np.ndarray, float]:\n fft_len = 2 ** _nextpow2(np.ceil(q * fs / fmin))\n \n if (fmax is not None) and (n_bins is None):\n n_bins = np.ceil(bins_per_octave * np.log2(fmax / fmin))\n freqs = fmin * 2.0 ** (np.r_[0:n_bins] / np.float(bins_per_octave))\n elif (fmax is None) and (n_bins is not None):\n freqs = fmin * 2.0 ** (np.r_[0:n_bins] / np.float(bins_per_octave))\n else:\n warnings.warn(\"If nmax is given, n_bins will be ignored\", SyntaxWarning)\n n_bins = np.ceil(bins_per_octave * np.log2(fmax / fmin))\n freqs = fmin * 2.0 ** (np.r_[0:n_bins] / np.float(bins_per_octave))\n if np.max(freqs) > fs / 2 and topbin_check:\n raise ValueError(f\"The top bin {np.max(freqs)} Hz has exceeded the Nyquist frequency, \\\n please reduce the `n_bins`\")\n \n kernel = np.zeros((int(n_bins), int(fft_len)), dtype=np.complex64)\n \n length = np.ceil(q * fs / freqs)\n for k in range(0, int(n_bins)):\n freq = freqs[k]\n l = np.ceil(q * fs / freq)\n \n if l % 2 == 1:\n start = int(np.ceil(fft_len / 2.0 - l / 2.0)) - 1\n else:\n start = int(np.ceil(fft_len / 2.0 - l / 2.0))\n\n sig = get_window(window, int(l), fftbins=True) * np.exp(\n np.r_[-l // 2:l // 2] * 1j * 2 * np.pi * freq / fs) / l\n \n if norm:\n kernel[k, start:start + int(l)] = sig / np.linalg.norm(sig, norm)\n else:\n kernel[k, start:start + int(l)] = sig\n return kernel, fft_len, length, freqs\n\ndef _nextpow2(a: float) -> int:\n return int(np.ceil(np.log2(a)))\n\n# Function to prepare cqt kernel\ndef prepare_cqt_kernel(\n sr=22050,\n hop_length=512,\n fmin=32.70,\n fmax=None,\n n_bins=84,\n bins_per_octave=12,\n norm=1,\n filter_scale=1,\n window=\"hann\"\n):\n q = float(filter_scale) / (2 ** (1 / bins_per_octave) - 1)\n print(q)\n return create_cqt_kernels(q, sr, fmin, n_bins, bins_per_octave, norm, window, fmax)\n# Function to create cqt image\ndef create_cqt_image(wave, hop_length=16):\n CQTs = []\n for i in range(3):\n x = wave[i]\n x = tf.expand_dims(tf.expand_dims(x, 0), 2)\n x = tf.pad(x, PADDING, \"REFLECT\")\n\n CQT_real = tf.nn.conv1d(x, CQT_KERNELS_REAL, stride=hop_length, padding=\"VALID\")\n CQT_imag = -tf.nn.conv1d(x, CQT_KERNELS_IMAG, stride=hop_length, padding=\"VALID\")\n CQT_real *= tf.math.sqrt(LENGTHS)\n CQT_imag *= tf.math.sqrt(LENGTHS)\n\n CQT = tf.math.sqrt(tf.pow(CQT_real, 2) + tf.pow(CQT_imag, 2))\n CQTs.append(CQT[0])\n return tf.stack(CQTs, axis=2)\n\nHOP_LENGTH = 64\ncqt_kernels, KERNEL_WIDTH, lengths, _ = prepare_cqt_kernel(\n sr=2048,\n hop_length=HOP_LENGTH,\n fmin=20,\n fmax=500,\n bins_per_octave=12)\nLENGTHS = tf.constant(lengths, dtype=tf.float32)\nCQT_KERNELS_REAL = tf.constant(np.swapaxes(cqt_kernels.real[:, np.newaxis, :], 0, 2))\nCQT_KERNELS_IMAG = tf.constant(np.swapaxes(cqt_kernels.imag[:, np.newaxis, :], 0, 2))\nPADDING = tf.constant([[0, 0],\n [KERNEL_WIDTH // 2, KERNEL_WIDTH // 2],\n [0, 0]])","metadata":{"id":"jnlgPmF0thwO","outputId":"4c6b0bb8-b226-4687-ba85-952d8542dc3a","execution":{"iopub.status.busy":"2021-09-17T18:44:47.097234Z","iopub.execute_input":"2021-09-17T18:44:47.097915Z","iopub.status.idle":"2021-09-17T18:44:47.332793Z","shell.execute_reply.started":"2021-09-17T18:44:47.097871Z","shell.execute_reply":"2021-09-17T18:44:47.331869Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

Reading TFrecords

","metadata":{}},{"cell_type":"code","source":"def read_labeled_tfrecord(example):\n tfrec_format = {\n \"wave\": tf.io.FixedLenFeature([], tf.string),\n \"wave_id\": tf.io.FixedLenFeature([], tf.string),\n \"target\": tf.io.FixedLenFeature([], tf.int64)\n }\n example = tf.io.parse_single_example(example, tfrec_format)\n return prepare_image(example[\"wave\"], config.image_size), tf.reshape(tf.cast(example[\"target\"], tf.float32), [1])\n\n\ndef read_unlabeled_tfrecord(example, return_image_id):\n tfrec_format = {\n \"wave\": tf.io.FixedLenFeature([], tf.string),\n \"wave_id\": tf.io.FixedLenFeature([], tf.string)\n }\n example = tf.io.parse_single_example(example, tfrec_format)\n return prepare_image(example[\"wave\"], config.image_size), example[\"wave_id\"] if return_image_id else 0\n\ndef count_data_items(fileids):\n return len(fileids) * 28000\n\ndef count_data_items_test(fileids):\n return len(fileids) * 22600\n\ndef prepare_image(wave, dim= config.image_size):\n wave = tf.reshape(tf.io.decode_raw(wave, tf.float64), (3, 4096))\n normalized_waves = []\n for i in range(3):\n normalized_wave = wave[i] / tf.math.reduce_max(wave[i])\n normalized_waves.append(normalized_wave)\n wave = tf.stack(normalized_waves)\n wave = tf.cast(wave, tf.float32)\n image = create_cqt_image(wave, HOP_LENGTH)\n image = tf.image.resize(image, size=(dim, dim))\n return tf.reshape(image, (dim, dim, 3))\n\ndef mixup(image, label, PROBABILITY = 1.0, AUG_BATCH= config.batch_size):\n # input image - is a batch of images of size [n,dim,dim,3] not a single image of [dim,dim,3]\n # output - a batch of images with mixup applied\n DIM = config.image_size\n \n imgs = []; labs = []\n for j in range(AUG_BATCH):\n # DO MIXUP WITH PROBABILITY DEFINED ABOVE\n P = tf.cast( tf.random.uniform([],0,1)<=PROBABILITY, tf.float32)\n # CHOOSE RANDOM\n k = tf.cast( tf.random.uniform([],0,AUG_BATCH),tf.int32)\n a = tf.random.uniform([],0,1)*P # this is beta dist with alpha=1.0\n # MAKE MIXUP IMAGE\n img1 = image[j,]\n img2 = image[k,]\n imgs.append((1-a)*img1 + a*img2)\n # MAKE CUTMIX LABEL\n lab1 = label[j,]\n lab2 = label[k,]\n labs.append((1-a)*lab1 + a*lab2)\n \n # RESHAPE HACK SO TPU COMPILER KNOWS SHAPE OF OUTPUT TENSOR (maybe use Python typing instead?)\n image2 = tf.reshape(tf.stack(imgs),(AUG_BATCH,DIM,DIM,3))\n label2 = tf.reshape(tf.stack(labs),(AUG_BATCH,))\n return image2,label2\n\ndef time_shift(img, shift= config.t_shift):\n if shift > 0:\n T = config.image_size\n P = tf.random.uniform([],0,1)\n SHIFT = tf.cast(T * P, tf.int32)\n return tf.concat([img[-SHIFT:], img[:-SHIFT]], axis=0)\n return img\n\n\ndef spector_shift(img, shift= config.s_shift):\n if shift > 0:\n T = config.image_size\n P = tf.random.uniform([],0,1)\n SHIFT = tf.cast(T * P, tf.int32)\n return tf.concat([img[:, -SHIFT:], img[:, :-SHIFT]], axis=1)\n return img\n \ndef rotate(img, angle= config.r_angle):\n if angle > 0:\n P = tf.random.uniform([],0,1)\n A = tf.cast(angle * P, tf.float32)\n return tfa.image.rotate(img, A)\n return img\n\ndef img_aug_f(img):\n img = time_shift(img)\n img = spector_shift(img)\n # img = rotate(img)\n return img\n\ndef imgs_aug_f(imgs, batch_size):\n _imgs = []\n DIM = config.image_size\n for j in range(batch_size):\n _imgs.append(img_aug_f(imgs[j]))\n return tf.reshape(tf.stack(_imgs),(batch_size,DIM,DIM,3))\n\ndef label_positive_shift(labels):\n return labels * config.label_positive_shift\n\ndef aug_f(imgs, labels, batch_size):\n imgs, label = mixup(imgs, labels, config.mix_up_p, batch_size)\n imgs = imgs_aug_f(imgs, batch_size)\n return imgs, label_positive_shift(label)\n","metadata":{"id":"7BBBxtrATx17","execution":{"iopub.status.busy":"2021-09-17T18:44:47.334696Z","iopub.execute_input":"2021-09-17T18:44:47.335021Z","iopub.status.idle":"2021-09-17T18:44:47.368141Z","shell.execute_reply.started":"2021-09-17T18:44:47.334977Z","shell.execute_reply":"2021-09-17T18:44:47.366423Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

📊 Dataset Creation

","metadata":{}},{"cell_type":"code","source":"def get_dataset(files, batch_size=16, repeat=False, shuffle=False, aug=True, labeled=True, return_image_ids=True):\n ds = tf.data.TFRecordDataset(files, num_parallel_reads=config.AUTO, compression_type=\"GZIP\")\n ds = ds.cache()\n\n if repeat:\n ds = ds.repeat()\n\n if shuffle:\n ds = ds.shuffle(1024 * 2)\n opt = tf.data.Options()\n opt.experimental_deterministic = False\n ds = ds.with_options(opt)\n\n if labeled:\n ds = ds.map(read_labeled_tfrecord, num_parallel_calls= config.AUTO)\n else:\n ds = ds.map(lambda example: read_unlabeled_tfrecord(example, return_image_ids), num_parallel_calls=config.AUTO)\n\n ds = ds.batch(config.batch_size)\n if aug:\n ds = ds.map(lambda x, y: aug_f(x, y, config.batch_size), num_parallel_calls= config.AUTO)\n ds = ds.prefetch(config.AUTO)\n return ds","metadata":{"id":"I1bF4jtgTx2H","execution":{"iopub.status.busy":"2021-09-17T18:44:47.369449Z","iopub.execute_input":"2021-09-17T18:44:47.369912Z","iopub.status.idle":"2021-09-17T18:44:47.388861Z","shell.execute_reply.started":"2021-09-17T18:44:47.369877Z","shell.execute_reply":"2021-09-17T18:44:47.387580Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

🧠 Modelling

","metadata":{}},{"cell_type":"code","source":"def create_model(backbone_model):\n x = GlobalAvgPool2D()(backbone_model.output)\n prediction_layer = Dense(1, activation = 'sigmoid')(x)\n \n model = Model(inputs = backbone_model.input , outputs = prediction_layer)\n \n return model\n","metadata":{"id":"a0jWSO2aTx2N","execution":{"iopub.status.busy":"2021-09-17T18:44:47.392229Z","iopub.execute_input":"2021-09-17T18:44:47.393829Z","iopub.status.idle":"2021-09-17T18:44:47.408833Z","shell.execute_reply.started":"2021-09-17T18:44:47.393774Z","shell.execute_reply":"2021-09-17T18:44:47.407718Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

🔄 KFold Prediction

","metadata":{}},{"cell_type":"code","source":"def get_predictions(dataset,fold_number):\n\n #create model \n print(\"creating model\")\n with strategy.scope():\n backbone_model = efn.EfficientNetB1(include_top= False, weights =\"imagenet\",input_shape=(config.image_size,config.image_size,3))\n model = create_model(backbone_model)\n\n #load weights\n print(\"loading model weights\")\n model.load_weights(f'../input/efficientnetb1-weights-g2net/result/eff_net_fold_{fold_number}.h5')\n \n pred = model.predict(dataset , verbose = 1 )\n pred = np.array(pred)\n\n return pred\n","metadata":{"id":"9tIdj4RarmJR","execution":{"iopub.status.busy":"2021-09-17T18:44:47.410802Z","iopub.execute_input":"2021-09-17T18:44:47.411077Z","iopub.status.idle":"2021-09-17T18:44:47.421105Z","shell.execute_reply.started":"2021-09-17T18:44:47.411040Z","shell.execute_reply":"2021-09-17T18:44:47.420156Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_data = get_dataset(test_files, batch_size= config.batch_size * 2, repeat=False, shuffle=False, aug=False, labeled=False, return_image_ids=True)\nids = np.array([id.numpy().decode(\"utf-8\") for _, id in tqdm(test_data.unbatch())]).flatten()","metadata":{"id":"2Sutwpd7ov-t","execution":{"iopub.status.busy":"2021-09-17T18:44:47.422868Z","iopub.execute_input":"2021-09-17T18:44:47.423162Z","iopub.status.idle":"2021-09-17T18:54:29.329645Z","shell.execute_reply.started":"2021-09-17T18:44:47.423112Z","shell.execute_reply":"2021-09-17T18:54:29.328450Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ds = get_dataset(test_files, batch_size= config.batch_size , repeat=False, shuffle=False, aug=False, labeled=False, return_image_ids=False)\nall_predictions = dict()\nfor fold in range(1,5):\n print(f'*********** PREDICTING FOR FOLD {fold} ***********')\n preds = get_predictions(ds,fold_number=fold)\n all_predictions[fold] = preds","metadata":{"id":"xg1ZVKgBqfIV","outputId":"d4fd0fb9-a998-4298-fa41-87feacc27407","execution":{"iopub.status.busy":"2021-09-17T18:54:29.331911Z","iopub.execute_input":"2021-09-17T18:54:29.332226Z","iopub.status.idle":"2021-09-17T19:03:34.532725Z","shell.execute_reply.started":"2021-09-17T18:54:29.332189Z","shell.execute_reply":"2021-09-17T19:03:34.531618Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sub_preds = np.mean([all_predictions[1].flatten(),all_predictions[2].flatten(),all_predictions[3].flatten(),\n all_predictions[4].flatten()],axis = 0)","metadata":{"id":"VQFMuqiX6d65","execution":{"iopub.status.busy":"2021-09-17T19:03:34.534220Z","iopub.execute_input":"2021-09-17T19:03:34.534498Z","iopub.status.idle":"2021-09-17T19:03:34.547235Z","shell.execute_reply.started":"2021-09-17T19:03:34.534467Z","shell.execute_reply":"2021-09-17T19:03:34.546161Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sub = pd.DataFrame()\nsub['id'] = ids\nsub['target'] = sub_preds\nsub","metadata":{"id":"DpUL04rh6hiy","outputId":"34c48ecd-c070-463a-fcde-bf36b9a5ba76","execution":{"iopub.status.busy":"2021-09-17T19:03:34.549112Z","iopub.execute_input":"2021-09-17T19:03:34.549470Z","iopub.status.idle":"2021-09-17T19:03:34.711415Z","shell.execute_reply.started":"2021-09-17T19:03:34.549427Z","shell.execute_reply":"2021-09-17T19:03:34.710399Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sub.to_csv('submission.csv', index=False)","metadata":{"id":"hkJIJxuhqHGW","execution":{"iopub.status.busy":"2021-09-17T19:03:34.712717Z","iopub.execute_input":"2021-09-17T19:03:34.712984Z","iopub.status.idle":"2021-09-17T19:03:35.444757Z","shell.execute_reply.started":"2021-09-17T19:03:34.712954Z","shell.execute_reply":"2021-09-17T19:03:35.443869Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" \n

📌 References

\n \n **Thank You Hidehisa Arai for providing such good kernels**\n * Hidehisa's [Training Notebook](https://www.kaggle.com/hidehisaarai1213/g2net-tf-on-the-fly-cqt-tpu-training) and [Inference Notebook](https://www.kaggle.com/hidehisaarai1213/g2net-tf-on-the-fly-cqt-tpu-inference)\n * Hidehisa's [Train Dataset](https://www.kaggle.com/hidehisaarai1213/g2net-waveform-tfrecords-train-0-4) and [Test Dataset](https://www.kaggle.com/hidehisaarai1213/g2net-waveform-tfrecords-test-0-4)","metadata":{}},{"cell_type":"markdown","source":"**Thanks for viewing, drop your suggestions down in the comments below. 🙂**","metadata":{}},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}