{
  "id": 11334,
  "title": "Does anyone use caffee? How could I produce a test result?",
  "url": "/competitions/datasciencebowl/discussion/11334",
  "author_name": "",
  "post_date": "2014-12-24T02:53:56.930Z",
  "votes": 2,
  "comment_count": 3,
  "views": 5755,
  "content": "<p>After learning dnn and try to use caffee, I find that there is no single document about how to produce a simple probability forecast for the test data set. All the tutorials are about how to train a dnn and how to calculate a validation score. Why there is no simple way&nbsp;to output the last layers's prediction? (and how)?</p>\n<p>Can anyone familiar with caffee give me an example? Thanks very much</p>\n<p>I have trained the network and could produce a validation score. But I don't know how to save the last layers output (softmax). I tried to use the tools/extract_features but the output is in lmdb format and it seems its hard to translate that to a txt file...</p>",
  "messages": [
    {
      "id": "61801",
      "postDate": "12/24/2014 02:53:56",
      "content": "<p>After learning dnn and try to use caffee, I find that there is no single document about how to produce a simple probability forecast for the test data set. All the tutorials are about how to train a dnn and how to calculate a validation score. Why there is no simple way&nbsp;to output the last layers's prediction? (and how)?</p>\n<p>Can anyone familiar with caffee give me an example? Thanks very much</p>\n<p>I have trained the network and could produce a validation score. But I don't know how to save the last layers output (softmax). I tried to use the tools/extract_features but the output is in lmdb format and it seems its hard to translate that to a txt file...</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "61806",
      "postDate": "12/24/2014 06:43:53",
      "content": "<p>Here is an example of how to use a trained Caffe model:</p>\n<p>http://nbviewer.ipython.org/github/BVLC/caffe/blob/master/examples/classification.ipynb</p>\n<p>It's in python.</p>\n<p>Also notice that you need a new .prototxt file when classifying images using an already trained model. By comparing the Caffe imagenet example prototxt files for training/validation and classifying, you can find their differences and know how to convert a training/validation prototxt file to classifying prototxt file.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "116151",
      "postDate": "04/22/2016 08:44:36",
      "content": "<p>link don't active(</p>",
      "rawMarkdown": "link don't active(",
      "votes": null
    },
    {
      "id": "157907",
      "postDate": "01/24/2017 09:51:53",
      "content": "<p>Try:\n<a href=\"http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb\">http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb</a></p>\n\n<p>Also:\n<a href=\"https://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy\">https://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy</a></p>",
      "rawMarkdown": "Try:\r\nhttp://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb\r\n\r\nAlso:\r\nhttps://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 61806,
      "author_name": "jinfengfeng",
      "author_url": "",
      "post_date": "12/24/2014 06:43:53",
      "content": "<p>Here is an example of how to use a trained Caffe model:</p>\n<p>http://nbviewer.ipython.org/github/BVLC/caffe/blob/master/examples/classification.ipynb</p>\n<p>It's in python.</p>\n<p>Also notice that you need a new .prototxt file when classifying images using an already trained model. By comparing the Caffe imagenet example prototxt files for training/validation and classifying, you can find their differences and know how to convert a training/validation prototxt file to classifying prototxt file.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 116151,
      "author_name": "danikkamilov",
      "author_url": "",
      "post_date": "04/22/2016 08:44:36",
      "content": "<p>link don't active(</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 157907,
      "author_name": "vektor",
      "author_url": "",
      "post_date": "01/24/2017 09:51:53",
      "content": "<p>Try:\n<a href=\"http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb\">http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb</a></p>\n\n<p>Also:\n<a href=\"https://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy\">https://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "61801": "",
    "61806": "",
    "116151": "link don't active(",
    "157907": "Try:\r\nhttp://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/00-classification.ipynb\r\n\r\nAlso:\r\nhttps://github.com/BVLC/caffe/wiki/Using-a-Trained-Network:-Deploy"
  },
  "source": "meta"
}