{"cells": [{"source": ["# Intro to Tensorflow \n", "\n", "Before engaging with this dataset I would recommend reading through https://www.kaggle.com/davids1992/data-visualization-and-investigation "], "metadata": {}, "cell_type": "markdown"}, {"source": ["## Let's start by Identifying our knowns...\n", "\n", "There are only 12 possible labels for the Test set: `yes`, `no`, `up`, `down`, `left`, `right`, `on`, `off`, `stop`, `go`, `silence`, `unknown`.\n", "\n", "The unknown label should be used for a command that is not one one of the first 10 labels or that is not silence."], "metadata": {}, "cell_type": "markdown"}, {"source": ["POSSIBLE_LABELS = 'yes no up down left right on off stop go silence unknown'.split()\n", "AUDIO_PATH = '../input/train/audio/'\n", "AUDIO_PATHS = {}\n", "for label in POSSIBLE_LABELS:\n", "    AUDIO_PATHS[label] = AUDIO_PATH + label\n", "print(AUDIO_PATHS)"], "outputs": [], "metadata": {}, "cell_type": "code", "execution_count": 10}, {"source": ["## First we need to turn our audio files into numbers \n", "then we can throw them into tensorflow "], "metadata": {}, "cell_type": "markdown"}, {"source": [], "outputs": [], "metadata": {"collapsed": true}, "cell_type": "code", "execution_count": null}], "nbformat": 4, "metadata": {"language_info": {"nbconvert_exporter": "python", "codemirror_mode": {"version": 3, "name": "ipython"}, "pygments_lexer": "ipython3", "mimetype": "text/x-python", "file_extension": ".py", "name": "python", "version": "3.6.3"}, "kernelspec": {"language": "python", "name": "python3", "display_name": "Python 3"}}, "nbformat_minor": 1}