{"cells":[{"metadata":{},"cell_type":"markdown","source":"> 데이터 분석 계획  \n1) 데이터 확인하기  \n2) 처리하기 적절한 형태로 데이터 가공  \n3) 모델링  \n4) 피드백 \n\n## 1) 데이터 확인 결과(test_simplified.csv, train_simplified)\n### 방법 :  \n23G에 달하는 대용량 data를 처리하기 위해 dask 패키지를 사용  \n### 결과 :   \ntest data의 column name) **'key_id  ', 'countrycode', 'drawing'**  \ntrain data의 column name) **'countrycode', 'drawing', 'key_id', 'recognized', 'timestamp', 'word'**\n\ntrain data는 category 별로 data가 각각의 csv 파일로 구성되어 있으며 최소 10만개 최대 30만개까지의 데이터를 포함한다.  \n나라의 문화에 따라 그림을 그리는 방식이나 그림의 모양새가 다를 것이라 추측했지만,  \n나라마다 일정 개수씩 수집된 것이 아니므로 특정 국가에 데이터가 편중된 것을 볼 수 있었다.  \n따라서, `countrycode를 고려하는 것은 무의미` 하다고 판단하였다.  \n\ndask 패키지를 이용하여 train data로 가상 data frame을 형성하였다.  \n이렇게 형성된 데이터에서 drawing과 word 열만 취하고 다시 random sampling 으로 일부 data만 가져왔다.  \n그런 다음, 6:2:2 의 비율로 train,valid, test data를 나누어 주었다.  \n\n## 2) 처리하기 적절한 형태로 데이터 가공\n### 방법 :  \ndrawing data가 string type으로 제공되었다.  \n하나의 stroke가 [x좌표의 sequce, y좌표의 squence]로 구성되어 있고, 이 stroke 들이 모여 하나의 그림을 만들어 낸다.  \n주어진 word에 따라 그림을 그리는 동안 AI가 정답을 맞출 수 있어야 하므로 필요한 data는 x와 y 좌표쌍 그리고 획(stroke)의 시작과 끝점이다.  \n주어진 drawing data에 json_loads를 적용시켜 list type으로 바꾸어 준 후, numpy의 stack 등을 이용하여 좌표쌍으로 구성한다.  \n\n## 3) 모델링\n### 방법 :  \n획들이 모여 하나의 그림이 된다. 이전에 그은 획의 위치가 달라지면 당연히 그림도 달라진다.  \n따라서, 이전에 그었던 획 정보들을 종합하여 그림이 속한 카테고리를 알아낼 수 있을 것이다.  \n과거의 정보를 반영하기 위해 다음과 같은 구조로 **Reccurent Neural Network(RNN)** 모델을 구성할 것이다. \n![rnn-model](https://www.tensorflow.org/images/quickdraw_model.png)\n\n먼저, 1D-convolution 을 적용시켜 feature를 풍부하게 하고  \n두 단계의 LSTM networks를 거친 후 340개의 category로 분류될 수 있도록 softmax 함수를 적용시켰다.  "},{"metadata":{},"cell_type":"markdown","source":"> import module"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import os\nfrom glob import glob # class별로 나뉘어 있는 train data를 받아오기 위해 glob module 을 사용\nimport pandas as pd\nimport numpy as np\nimport dask.dataframe as dd # 대용량 파일을 읽기위한 패키지\nfrom dask.diagnostics import ProgressBar\nimport matplotlib.pyplot as plot\nimport json \n\npbar = ProgressBar()\npbar.register()\n\nprint(os.listdir(\"../input\"))","execution_count":2,"outputs":[{"output_type":"stream","text":"['sample_submission.csv', 'test_simplified.csv', 'test_raw.csv', 'train_simplified']\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow as tf\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.preprocessing import OneHotEncoder\nimport keras \nfrom keras.utils import to_categorical\nfrom keras.preprocessing.sequence import pad_sequences\nfrom keras.models import Sequential\n","execution_count":3,"outputs":[{"output_type":"stream","text":"Using TensorFlow backend.\n","name":"stderr"}]},{"metadata":{},"cell_type":"markdown","source":"## 데이터 불러오기"},{"metadata":{"trusted":true},"cell_type":"code","source":"base_dir = os.path.join('..','input')\ntest_path = os.path.join(base_dir,'test_simplified.csv')\ntrain_paths = glob(os.path.join(base_dir, 'train_simplified', '*.csv'))\n# class별로 나뉘어있는 train data를 받아오기 위해 glob module 을 사용\n\ntest_data = pd.read_csv(test_path)\nprint(\"test_data columns: \",test_data.columns)\ntrain_data = pd.read_csv(train_paths[0])\nprint(\"train_data columns: \",train_data.columns)\nprint(\"draw data type: \", type(train_data['drawing'][0]))","execution_count":4,"outputs":[{"output_type":"stream","text":"test_data columns:  Index(['key_id', 'countrycode', 'drawing'], dtype='object')\ntrain_data columns:  Index(['countrycode', 'drawing', 'key_id', 'recognized', 'timestamp', 'word'], dtype='object')\ndraw data type:  <class 'str'>\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.head()","execution_count":5,"outputs":[{"output_type":"execute_result","execution_count":5,"data":{"text/plain":"  countrycode  ...   word\n0          GB  ...   leaf\n1          PK  ...   leaf\n2          US  ...   leaf\n3          AE  ...   leaf\n4          CA  ...   leaf\n\n[5 rows x 6 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>countrycode</th>\n      <th>drawing</th>\n      <th>key_id</th>\n      <th>recognized</th>\n      <th>timestamp</th>\n      <th>word</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>GB</td>\n      <td>[[[0, 52, 79, 95, 119, 154, 175, 232, 253, 254...</td>\n      <td>5308841862365184</td>\n      <td>True</td>\n      <td>2017-03-16 08:17:09.868490</td>\n      <td>leaf</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>PK</td>\n      <td>[[[56, 56], [67, 67]], [[49, 51, 44, 44, 48, 4...</td>\n      <td>6346088359395328</td>\n      <td>True</td>\n      <td>2017-03-15 20:27:48.393710</td>\n      <td>leaf</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>US</td>\n      <td>[[[0, 60, 124], [255, 162, 92]], [[33, 57, 87,...</td>\n      <td>5804594418417664</td>\n      <td>False</td>\n      <td>2017-03-01 21:03:23.722220</td>\n      <td>leaf</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>AE</td>\n      <td>[[[0, 47, 105], [167, 103, 60]], [[85, 81, 81,...</td>\n      <td>5531143698907136</td>\n      <td>True</td>\n      <td>2017-03-29 18:43:11.841340</td>\n      <td>leaf</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>CA</td>\n      <td>[[[103, 107], [255, 177]], [[112, 103, 83, 33]...</td>\n      <td>6632150193405952</td>\n      <td>True</td>\n      <td>2017-03-01 00:22:10.294980</td>\n      <td>leaf</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"## drawing data-> (x,y,시작or끝)으로 데이터 가공  \n궁금한 점 : stroke 마다 점의 수가 다른데 padding 해주어야 할까 ? \n=> 해주어야 한다. 길이가 일정한 입력을 만들어주기 위해서 "},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"# drawing data를 [number of points,3] 크기의 tensor로 바꾸어주는 함수\ndef make_stroke2tensor(raw_strokes):\n    # string type의 drawing data를 list로 변환\n    stroke_lst = json.loads(raw_strokes)\n    # storke_lst = (coord_x list,coor_y list) 들로 구성\n    # -> 쌍이 되는 (x,y,stroke id)들로 이루어진 list를 만듦\n    stroke_coords = [(x,y,i) for i,(x_lst,y_lst) in enumerate(stroke_lst) \n                  for x,y in zip(x_lst,y_lst)]\n    # (x,y,index)들의 리스트인 그림 데이터 하나를 원소로 가지는 리스트로 구성\n    stroke_coords = np.stack(stroke_coords)\n    \n    # 획의 시작과 끝의 정보 저장 \n    stroke_coords[:,2] = [1]+np.diff(stroke_coords[:,2]).tolist()\n    stroke_coords[:,2] += 1\n    \n    # return stroke_coords\n    \n    return pad_sequences(stroke_coords.swapaxes(0, 1), \n                         maxlen=200, \n                         padding='post').swapaxes(0, 1)\n\ntrain_data.drawing = train_data.drawing.map(make_stroke2tensor)\n","execution_count":6,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# 임의의 train data file 을 살펴보니, train_data의 shape은 약 12만개의 데이터가 6개의 정보를 담고 있는 형태이고,\n# drawing data는 1000개의 점으로 이루어진 data이다.\nprint(train_data.shape)\nprint(train_data.drawing.shape)\nprint(train_data.drawing[0].shape)","execution_count":7,"outputs":[{"output_type":"stream","text":"(125571, 6)\n(125571,)\n(200, 3)\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X = np.stack(train_data.drawing,0)\nY = train_data.drawing\nY = np.stack(Y,0)\nprint(Y)","execution_count":8,"outputs":[{"output_type":"stream","text":"[[[  0 203   2]\n  [ 52 170   1]\n  [ 79  82   1]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]\n\n [[ 56  67   2]\n  [ 56  67   1]\n  [ 49 255   2]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]\n\n [[  0 255   2]\n  [ 60 162   1]\n  [124  92   1]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]\n\n ...\n\n [[  0 252   2]\n  [ 14 229   1]\n  [ 26 218   1]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]\n\n [[ 82 249   2]\n  [ 82 195   1]\n  [ 89   0   2]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]\n\n [[ 45   8   2]\n  [ 16  47   1]\n  [  1  86   1]\n  ...\n  [  0   0   0]\n  [  0   0   0]\n  [  0   0   0]]]\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X.shape\nY.shape","execution_count":9,"outputs":[{"output_type":"execute_result","execution_count":9,"data":{"text/plain":"(125571, 200, 3)"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# train_data.drawing # [[drawing data 나열:[x,y,startORend_info]]]","execution_count":10,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## data 구성하기"},{"metadata":{"trusted":true},"cell_type":"code","source":"whole_train_data = dd.read_csv(train_paths)","execution_count":11,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# whole_train_data.count().compute()","execution_count":12,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"a = whole_train_data[['drawing','word']].sample(frac = 0.008)","execution_count":13,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data,valid_data,train_test_data =a. random_split([0.60,0.20,0.20])","execution_count":14,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.drawing.head()","execution_count":15,"outputs":[{"output_type":"stream","text":"[########################################] | 100% Completed |  0.8s\n","name":"stdout"},{"output_type":"execute_result","execution_count":15,"data":{"text/plain":"43216    [[[0, 48, 86], [131, 117, 99]], [[87, 85, 87, ...\n99217    [[[39, 25, 21, 16, 1, 1, 11, 24, 56, 100, 104,...\n28874    [[[69, 47, 25, 15, 15, 28, 60, 65, 90, 100, 10...\n36172    [[[173, 109, 63, 49, 54], [255, 202, 136, 85, ...\n63635    [[[53, 54, 49], [221, 255, 226]], [[58, 29, 13...\nName: drawing, dtype: object"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_word = train_data['word']\ntrain_drawing = train_data.drawing\ntrain_drawing = train_drawing.map(make_stroke2tensor,meta=('drawing', int))\n\nvalid_word = valid_data['word']\nvalid_drawing = valid_data.drawing\nvalid_drawing = valid_drawing.map(make_stroke2tensor,meta=('drawing', int))\n\ntrain_test_word = train_test_data['word']\ntrain_test_drawing = train_test_data.drawing\ntrain_test_drawing = train_test_drawing.map(make_stroke2tensor,meta=('drawing',int))\n","execution_count":16,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# word 형태의 category를 one-hot encoding으로 분류하기 위함 \nword_encoder = LabelEncoder()\nword_encoder.fit(train_word)\n\ntrain_word = to_categorical(word_encoder.transform(train_word.values))\nvalid_word = to_categorical(word_encoder.transform(valid_word.values))\ntrain_test_word = to_categorical(word_encoder.transform(train_test_word.values))\n\n","execution_count":17,"outputs":[{"output_type":"stream","text":"[########################################] | 100% Completed |  5min 22.2s\n[########################################] | 100% Completed |  5min 19.4s\n[########################################] | 100% Completed |  5min 15.8s\n[########################################] | 100% Completed |  5min 14.9s\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_drawing =np.stack(np.array(train_drawing),0)\nvalid_drawing =np.stack(np.array(valid_drawing),0)\ntrain_test_drawing =np.stack(np.array(train_test_drawing),0)\n","execution_count":18,"outputs":[{"output_type":"stream","text":"[########################################] | 100% Completed |  6min 31.5s\n[########################################] | 100% Completed |  5min 30.9s\n[########################################] | 100% Completed |  5min 42.3s\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(train_drawing.shape)\nprint(train_drawing.dtype)\nprint(train_word.shape)\nprint(train_word.dtype)","execution_count":20,"outputs":[{"output_type":"stream","text":"(238561, 200, 3)\nint32\n(238561, 340)\nfloat32\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"def preds2catids(predictions):\n    return pd.DataFrame(np.argsort(-predictions, axis=1)[:, :3], columns=['a', 'b', 'c'])\n\ndef top_3_accuracy(y_true, y_pred):\n    return top_k_categorical_accuracy(y_true, y_pred, k=3)","execution_count":21,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"learning_rate = 1e-3\nnum_epochs = 15\nbatch_size = 2048\nnum_display = 100","execution_count":22,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.layers import CuDNNLSTM as LSTM\nfrom keras.layers import BatchNormalization,Conv1D, Dense, Dropout\n# keras 순차모델 생성\nfrom keras.metrics import top_k_categorical_accuracy\n","execution_count":23,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = Sequential()","execution_count":24,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.add(BatchNormalization(input_shape = (None,)+train_drawing.shape[2:]))","execution_count":26,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\nInstructions for updating:\nColocations handled automatically by placer.\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.add(Conv1D(64,(3,)))\nmodel.add(Dropout(rate = 0.8))\nmodel.add(Conv1D(128,(3,)))\nmodel.add(Dropout(rate = 0.8))\nmodel.add(Conv1D(256,(3,)))\nmodel.add(Dropout(rate = 0.8))","execution_count":27,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\nInstructions for updating:\nPlease use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.add(LSTM(128,return_sequences = True))\nmodel.add(Dropout(rate = 0.8))","execution_count":28,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.add(LSTM(256,return_sequences = False))\nmodel.add(Dropout(rate = 0.8))","execution_count":29,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.add(Dense(340,activation = 'softmax'))","execution_count":30,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.compile(optimizer = 'adam',\n             loss = 'categorical_crossentropy',\n             metrics = ['categorical_accuracy',top_3_accuracy])\n","execution_count":31,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.summary()","execution_count":32,"outputs":[{"output_type":"stream","text":"_________________________________________________________________\nLayer (type)                 Output Shape              Param #   \n=================================================================\nbatch_normalization_1 (Batch (None, None, 3)           12        \n_________________________________________________________________\nconv1d_1 (Conv1D)            (None, None, 64)          640       \n_________________________________________________________________\ndropout_1 (Dropout)          (None, None, 64)          0         \n_________________________________________________________________\nconv1d_2 (Conv1D)            (None, None, 128)         24704     \n_________________________________________________________________\ndropout_2 (Dropout)          (None, None, 128)         0         \n_________________________________________________________________\nconv1d_3 (Conv1D)            (None, None, 256)         98560     \n_________________________________________________________________\ndropout_3 (Dropout)          (None, None, 256)         0         \n_________________________________________________________________\ncu_dnnlstm_1 (CuDNNLSTM)     (None, None, 128)         197632    \n_________________________________________________________________\ndropout_4 (Dropout)          (None, None, 128)         0         \n_________________________________________________________________\ncu_dnnlstm_2 (CuDNNLSTM)     (None, 256)               395264    \n_________________________________________________________________\ndropout_5 (Dropout)          (None, 256)               0         \n_________________________________________________________________\ndense_1 (Dense)              (None, 340)               87380     \n=================================================================\nTotal params: 804,192\nTrainable params: 804,186\nNon-trainable params: 6\n_________________________________________________________________\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.callbacks import ModelCheckpoint, LearningRateScheduler, EarlyStopping, ReduceLROnPlateau","execution_count":33,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"weight_path=\"{}_weights.best.hdf5\".format('model')\n\ncheckpoint = ModelCheckpoint(weight_path, monitor='val_loss', verbose=1, \n                             save_best_only=True, mode='min', save_weights_only = True)\n\n\nreduceLROnPlat = ReduceLROnPlateau(monitor='val_loss', factor=0.8, patience=10, \n                                   verbose=1, mode='auto', epsilon=0.0001, cooldown=5, min_lr=0.0001)\nearly = EarlyStopping(monitor=\"val_loss\", \n                      mode=\"min\", \n                      patience=5) # probably needs to be more patient, but kaggle time is limited\ncallbacks_list = [checkpoint, early, reduceLROnPlat]","execution_count":34,"outputs":[{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/keras/callbacks.py:1065: UserWarning: `epsilon` argument is deprecated and will be removed, use `min_delta` instead.\n  warnings.warn('`epsilon` argument is deprecated and '\n","name":"stderr"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"from IPython.display import clear_output\nmodel.fit(train_drawing, train_word,\n                      validation_data = (valid_drawing, valid_word), \n                      batch_size = batch_size,\n                      epochs = 50,\n                      callbacks = callbacks_list)\nclear_output()","execution_count":39,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.load_weights(weight_path)\nlstm_results = model.evaluate(train_test_drawing, train_test_word, batch_size = 2048)\nprint('Accuracy: %2.1f%%, Top 3 Accuracy %2.1f%%' % (100*lstm_results[1], 100*lstm_results[2]))","execution_count":40,"outputs":[{"output_type":"stream","text":"79758/79758 [==============================] - 5s 65us/step\nAccuracy: 35.1%, Top 3 Accuracy 54.5%\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"### Submission"},{"metadata":{"trusted":true},"cell_type":"code","source":"sub_drawing = test_data['drawing'].map(make_stroke2tensor)\nsub_drawing = np.stack(sub_drawing.values,0)\nsub_pred = model.predict(sub_drawing,verbose = True, batch_size = 2048)","execution_count":42,"outputs":[{"output_type":"stream","text":"112199/112199 [==============================] - 8s 67us/step\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"top_3_pred = [word_encoder.classes_[np.argsort(-1*c_pred)[:3]] for c_pred in sub_pred]","execution_count":43,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"top_3_pred = [' '.join([col.replace(' ', '_') for col in row]) for row in top_3_pred]\ntop_3_pred[:3]","execution_count":44,"outputs":[{"output_type":"execute_result","execution_count":44,"data":{"text/plain":"['radio stereo peas', 'hockey_puck bathtub bottlecap', 'camel hand castle']"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_data['word'] = top_3_pred\ntest_data[['key_id', 'word']].to_csv('submission.csv', index=False)","execution_count":48,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Show some predictions on the submission dataset"},{"metadata":{"trusted":true},"cell_type":"code","source":"import matplotlib.pyplot as plt","execution_count":46,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, m_axs = plt.subplots(3,3, figsize = (16, 16))\nrand_idxs = np.random.choice(range(sub_drawing.shape[0]), size = 9)\nfor c_id, c_ax in zip(rand_idxs, m_axs.flatten()):\n    test_arr = sub_drawing[c_id]\n    test_arr = test_arr[test_arr[:,2]>0, :] # only keep valid points\n    lab_idx = np.cumsum(test_arr[:,2]-1)\n    for i in np.unique(lab_idx):\n        c_ax.plot(test_arr[lab_idx==i,0], \n                np.max(test_arr[:,1])-test_arr[lab_idx==i,1], '.-')\n    c_ax.axis('off')\n    c_ax.set_title(top_3_pred[c_id])","execution_count":47,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1152x1152 with 9 Axes>","image/png":"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\n"},"metadata":{}}]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}