{
  "id": 45504,
  "title": "Create a set of inputs to frozen TensorFlow GraphDef file for Raspberry Pi 3",
  "url": "/competitions/tensorflow-speech-recognition-challenge/discussion/45504",
  "author_name": "Huynh",
  "post_date": "2017-12-12T09:18:56.710000",
  "votes": 1,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi,\nI have problem with create the set of inputs as competition's Prizes: </p>\n\n<blockquote>\n  <p>decoded_sample_data:0, taking a [16000, 1] float tensor as input, representing the audio PCM-encoded data. </p>\n  \n  <p>decoded_sample_data:1, taking a scalar [] int32 tensor as input,  representing the sample rate, which must be the value 16000.</p>\n</blockquote>\n\n<p>I have create 2 placeholder as follow:</p>\n\n<p>sample_placeholder = tf.placeholder(dtype=tf.float32, shape=[16000,1],name='decoded_sample_data')\nsr_placeholder = tf.placeholder(dtype=tf.int32, name='decoded_sample_data')</p>\n\n<p>but the name is different. How can I create this kind of inputs? Thanks!</p>",
  "messages": [
    {
      "id": 256569,
      "postDate": "2017-12-12T09:18:56.710Z",
      "content": "<p>Hi,\nI have problem with create the set of inputs as competition's Prizes: </p>\n\n<blockquote>\n  <p>decoded_sample_data:0, taking a [16000, 1] float tensor as input, representing the audio PCM-encoded data. </p>\n  \n  <p>decoded_sample_data:1, taking a scalar [] int32 tensor as input,  representing the sample rate, which must be the value 16000.</p>\n</blockquote>\n\n<p>I have create 2 placeholder as follow:</p>\n\n<p>sample_placeholder = tf.placeholder(dtype=tf.float32, shape=[16000,1],name='decoded_sample_data')\nsr_placeholder = tf.placeholder(dtype=tf.int32, name='decoded_sample_data')</p>\n\n<p>but the name is different. How can I create this kind of inputs? Thanks!</p>",
      "rawMarkdown": "Hi,\nI have problem with create the set of inputs as competition's Prizes: \n\n&gt; decoded_sample_data:0, taking a [16000, 1] float tensor as input, representing the audio PCM-encoded data. \n\n&gt; decoded_sample_data:1, taking a scalar [] int32 tensor as input,  representing the sample rate, which must be the value 16000.\n\nI have create 2 placeholder as follow:\n\nsample_placeholder = tf.placeholder(dtype=tf.float32, shape=[16000,1],name='decoded_sample_data')\nsr_placeholder = tf.placeholder(dtype=tf.int32, name='decoded_sample_data')\n\nbut the name is different. How can I create this kind of inputs? Thanks!\n"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "256569": "Hi,\nI have problem with create the set of inputs as competition's Prizes: \n\n&gt; decoded_sample_data:0, taking a [16000, 1] float tensor as input, representing the audio PCM-encoded data. \n\n&gt; decoded_sample_data:1, taking a scalar [] int32 tensor as input,  representing the sample rate, which must be the value 16000.\n\nI have create 2 placeholder as follow:\n\nsample_placeholder = tf.placeholder(dtype=tf.float32, shape=[16000,1],name='decoded_sample_data')\nsr_placeholder = tf.placeholder(dtype=tf.int32, name='decoded_sample_data')\n\nbut the name is different. How can I create this kind of inputs? Thanks!\n"
  }
}