# 11-785 Homework 1 Part 2 - Fall 19 (LATE) — Data

来源：https://www.kaggle.com/competitions/11-785-hw1-fall2019-late/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description

<h1>File descriptions</h1>

<p>Data files are available for download from the original competition page.</p>

<h2>File descriptions</h2>

<ul>
<li><strong>train.npy</strong> - training features</li>
<li><strong>train_labels.npy</strong> - training labels</li>
<li><strong>dev.npy</strong> - development/validation features</li>
<li><strong>dev_labels.npy</strong> - development/validation labels</li>
<li><strong>test.npy</strong> - test features</li>
<li><strong>sample_submission.csv</strong> - sample submission in correct format</li>
</ul>

<h2>Feature Files</h2>

<p><code>[train|dev|test].npy</code> contain a numpy object array of shape <code>[utterances]</code>. Each utterance is a float32 ndarray of shape <code>[time, frequency]</code>, where time is the length of the utterance. Frequency dimension is always 40 but time dimension is of variable length.</p>

<h2>Label Files</h2>

<p><code>[train|dev]_labels.npy</code> contain a numpy object array of shape <code>[utterances]</code>. Each element in the array is an int32 array of shape [time] and provides the phoneme state label for each frame. There are 138 distinct labels [0-137], one for each subphoneme.</p>

<h2>Loader</h2>

<p>We have provided some starter code to help with loading the data. You can find the gist <a href="https://gist.github.com/rbrigden/d611ef4f2eccddb5581e0ac617ce38eb" rel="nofollow">here</a></p>