{
  "id": 20792,
  "title": "Tutorials for fine tuning(python).",
  "url": "/competitions/state-farm-distracted-driver-detection/discussion/20792",
  "author_name": "Vladimir Iglovikov",
  "post_date": "2016-05-07T17:18:21.483000",
  "votes": 0,
  "comment_count": 0,
  "views": 704,
  "content": "<p>Hello everyone,</p>\n\n<p>There are number of different packages for neural networks and number of pretrained neural networks in the internet. Idea for fine tuning is great and straightforward, but syntax is different. Of course I can go through someones code that uses fine tuning, and many people already shared their approach, but if there are some tutorials (ipython notebooks) with extensive commenting, it can make learning process slightly faster. </p>\n\n<p>I know that caffe has such a <a href=\"http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/02-fine-tuning.ipynb\">tutorial</a>. I adapted it for this competition but there is still some bug, and I will post here after I will fix issues in it. But it would be great if you, guys, would share tutorials or may be blog posts about other packages(I would not mind some other fine tuning example for caffe). Nervana? Keras? Nolean/Lasagne? Graphlab?</p>\n\n<p>Vladimir</p>",
  "messages": [
    {
      "id": 119153,
      "postDate": "2016-05-07T17:18:21.483Z",
      "content": "<p>Hello everyone,</p>\n\n<p>There are number of different packages for neural networks and number of pretrained neural networks in the internet. Idea for fine tuning is great and straightforward, but syntax is different. Of course I can go through someones code that uses fine tuning, and many people already shared their approach, but if there are some tutorials (ipython notebooks) with extensive commenting, it can make learning process slightly faster. </p>\n\n<p>I know that caffe has such a <a href=\"http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/02-fine-tuning.ipynb\">tutorial</a>. I adapted it for this competition but there is still some bug, and I will post here after I will fix issues in it. But it would be great if you, guys, would share tutorials or may be blog posts about other packages(I would not mind some other fine tuning example for caffe). Nervana? Keras? Nolean/Lasagne? Graphlab?</p>\n\n<p>Vladimir</p>",
      "rawMarkdown": "Hello everyone,\r\n\r\nThere are number of different packages for neural networks and number of pretrained neural networks in the internet. Idea for fine tuning is great and straightforward, but syntax is different. Of course I can go through someones code that uses fine tuning, and many people already shared their approach, but if there are some tutorials (ipython notebooks) with extensive commenting, it can make learning process slightly faster. \r\n\r\nI know that caffe has such a [tutorial][1]. I adapted it for this competition but there is still some bug, and I will post here after I will fix issues in it. But it would be great if you, guys, would share tutorials or may be blog posts about other packages(I would not mind some other fine tuning example for caffe). Nervana? Keras? Nolean/Lasagne? Graphlab?\r\n\r\nVladimir\r\n\r\n\r\n  [1]: http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/02-fine-tuning.ipynb"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "119153": "Hello everyone,\r\n\r\nThere are number of different packages for neural networks and number of pretrained neural networks in the internet. Idea for fine tuning is great and straightforward, but syntax is different. Of course I can go through someones code that uses fine tuning, and many people already shared their approach, but if there are some tutorials (ipython notebooks) with extensive commenting, it can make learning process slightly faster. \r\n\r\nI know that caffe has such a [tutorial][1]. I adapted it for this competition but there is still some bug, and I will post here after I will fix issues in it. But it would be great if you, guys, would share tutorials or may be blog posts about other packages(I would not mind some other fine tuning example for caffe). Nervana? Keras? Nolean/Lasagne? Graphlab?\r\n\r\nVladimir\r\n\r\n\r\n  [1]: http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/02-fine-tuning.ipynb"
  }
}