{
  "id": 21559,
  "title": "Running Keras on CPU",
  "url": "/competitions/state-farm-distracted-driver-detection/discussion/21559",
  "author_name": "",
  "post_date": "2016-06-09T22:11:06.427Z",
  "votes": null,
  "comment_count": 1,
  "views": 1106,
  "content": "<p>Hi,</p>\n\n<p>I just realize that my desktop (window 7) only has Intel(R) HD Graphics 4600, and not NIVIDIA, so I cannot run theano on GPU with the CUDA 7.5 package. Is there a way to run it on the Intel HD graphics card that I have, or speed up the model.fit process while running on cpu? Thanks!</p>",
  "messages": [
    {
      "id": "123130",
      "postDate": "06/09/2016 22:11:06",
      "content": "<p>Hi,</p>\n\n<p>I just realize that my desktop (window 7) only has Intel(R) HD Graphics 4600, and not NIVIDIA, so I cannot run theano on GPU with the CUDA 7.5 package. Is there a way to run it on the Intel HD graphics card that I have, or speed up the model.fit process while running on cpu? Thanks!</p>",
      "rawMarkdown": "Hi,\r\n\r\n I just realize that my desktop (window 7) only has Intel(R) HD Graphics 4600, and not NIVIDIA, so I cannot run theano on GPU with the CUDA 7.5 package. Is there a way to run it on the Intel HD graphics card that I have, or speed up the model.fit process while running on cpu? Thanks!",
      "votes": null
    },
    {
      "id": "123747",
      "postDate": "06/13/2016 20:49:42",
      "content": "<p>To my knowledge, those Deep Learning Libraries only support CPUs and NVIDIA GPUs. It is a waste of time to run scripts in CPU mode. NVIDIA GPUs could accelerate computing significantly. FYI, the g2.2xlarge instance on AWS offers a decent option for you :-)</p>",
      "rawMarkdown": "To my knowledge, those Deep Learning Libraries only support CPUs and NVIDIA GPUs. It is a waste of time to run scripts in CPU mode. NVIDIA GPUs could accelerate computing significantly. FYI, the g2.2xlarge instance on AWS offers a decent option for you :-)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 123747,
      "author_name": "xingyang",
      "author_url": "",
      "post_date": "06/13/2016 20:49:42",
      "content": "<p>To my knowledge, those Deep Learning Libraries only support CPUs and NVIDIA GPUs. It is a waste of time to run scripts in CPU mode. NVIDIA GPUs could accelerate computing significantly. FYI, the g2.2xlarge instance on AWS offers a decent option for you :-)</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "123130": "Hi,\r\n\r\n I just realize that my desktop (window 7) only has Intel(R) HD Graphics 4600, and not NIVIDIA, so I cannot run theano on GPU with the CUDA 7.5 package. Is there a way to run it on the Intel HD graphics card that I have, or speed up the model.fit process while running on cpu? Thanks!",
    "123747": "To my knowledge, those Deep Learning Libraries only support CPUs and NVIDIA GPUs. It is a waste of time to run scripts in CPU mode. NVIDIA GPUs could accelerate computing significantly. FYI, the g2.2xlarge instance on AWS offers a decent option for you :-)"
  },
  "source": "meta"
}