{"cells":[{"metadata":{},"cell_type":"markdown","source":"## A quick look at the data\nWe will quickly explore what the data looks like and how we may proceed."},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np \nimport pandas as pd\n%pylab inline\n\n# Input data files are available in the \"../input/\" directory.\nimport os\nprint(os.listdir(\"../input\"))\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Explore training data\n- The file is large. Let's start with reading a small number rows first."},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"# read the training data\n# the data is large. Read a small number of rows for now\ntrain_data = pd.read_csv('../input/train.csv', nrows=100000)\ntrain_data.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Let's plot part of the data to see what it looks like."},{"metadata":{"trusted":true},"cell_type":"code","source":"\nfig, ax = plt.subplots(1,4,figsize=(13,4))\nax[0].plot(train_data.time_to_failure.values)\nax[0].set_xlabel('Index'); ax[0].set_ylabel('time to failure')\nax[1].plot(train_data.acoustic_data.values)\nax[1].set_xlabel('Index'); ax[1].set_ylabel('acoustic data')\nax[2].plot(np.diff(train_data.time_to_failure.values))\nax[2].set_xlabel('Index'); ax[2].set_ylabel('step of time_to_failure')\nax[3].plot(train_data.acoustic_data.values, train_data.time_to_failure.values, 'o', alpha=0.1)\nax[3].set_xlabel('acoustic data'); ax[3].set_ylabel('time to failure')\nplt.tight_layout(pad=2)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"The `time_to_failure` does not appear to change from the table. It is in fact changing but very slowly by roughly 1 ns, then every 4096 point it jumps by 1 ms.\n\nThe 4096 segments could be related to the way the instrument takes the measurements. We can plot 3 segments to see what's going on.\n\n"},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots(1,4,figsize=(13,4))\nnplt = 4096*3\nax[0].plot(train_data.time_to_failure.values[:nplt])\nax[0].set_xlabel('Index'); ax[0].set_ylabel('time to failure')\nax[1].plot(train_data.acoustic_data.values[:nplt])\nax[1].set_xlabel('Index'); ax[1].set_ylabel('acoustic data')\nax[2].plot(np.diff(train_data.time_to_failure.values[:nplt]))\nax[2].set_xlabel('Index'); ax[2].set_ylabel('step of time_to_failure')\nax[3].plot(train_data.acoustic_data.values[:nplt], train_data.time_to_failure.values[:nplt], 'o', alpha=0.1)\nax[3].set_xlabel('acoustic data'); ax[3].set_ylabel('time to failure')\nplt.tight_layout(pad=2)","execution_count":11,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 936x288 with 4 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"So it looks like the acoustic data changes within the 4096 segments, so there is useful information that we want to use.\nThe `time_to_failure` is still decreasing, so we haven't seen a quake yet. Let's load more data\n\n### Loading longer segments\n"},{"metadata":{"trusted":true},"cell_type":"code","source":"# read the training data\n# the data is large. Read a small number of rows for now\ntrain_data_long = pd.read_csv('../input/train.csv', nrows=10000000)\ntrain_data_long.head()","execution_count":12,"outputs":[{"output_type":"execute_result","execution_count":12,"data":{"text/plain":"   acoustic_data  time_to_failure\n0             12           1.4691\n1              6           1.4691\n2              8           1.4691\n3              5           1.4691\n4              8           1.4691","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>acoustic_data</th>\n      <th>time_to_failure</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>12</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>6</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>8</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>5</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>8</td>\n      <td>1.4691</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# plot every 500 points, so we explore the data quickly\nfig, ax = plt.subplots(1,3,figsize=(13,4))\nax[0].plot(train_data_long.time_to_failure.values[::500])\nax[0].set_xlabel('Index'); ax[0].set_ylabel('time to failure')\nax[1].plot(train_data_long.acoustic_data.values[::500])\nax[1].set_xlabel('Index'); ax[1].set_ylabel('acoustic data')\nax[2].plot(train_data_long.acoustic_data.values[::500], train_data_long.time_to_failure.values[::500], 'o', alpha=0.1)\nax[2].set_xlabel('acoustic data'); ax[2].set_ylabel('time to failure')\nplt.tight_layout(pad=2)","execution_count":13,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 936x288 with 3 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"Now we see the actual quakes as a step jump in the `time_to_failure` value after reaching 0. The `acoustic_data` appears to vary more just before the quake happen. This is presumably one of the indications that a quake is about to happend and that our model needs to capture."},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"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}