)]}'
{
  "commit": "fcfb1c1979b63c77b6fa6186b622dc08253f9791",
  "tree": "0b0586de09ee4b0ea21f961710db7cd64b2604d7",
  "parents": [
    "529fc68df280663e1c7d145a70ef3ddb20c62e36"
  ],
  "author": {
    "name": "Fritz Obermeyer",
    "email": "fritzo@uber.com",
    "time": "Thu Aug 23 08:06:43 2018 -0700"
  },
  "committer": {
    "name": "Facebook Github Bot",
    "email": "facebook-github-bot@users.noreply.github.com",
    "time": "Thu Aug 23 08:09:49 2018 -0700"
  },
  "message": "Make more distributions jittable\n\nSummary:\nThis uses zou3519\u0027s new `torch.broadcast_tensors()` #10075 to make `Categorical.log_prob()` and the `*Normal.__init__()` methods jittable. Previously `.log_prob()` was failing due to calls to `torch._C.infer_size()` with errors like\n```\n    def log_prob(self, value):\n        if self._validate_args:\n            self._validate_sample(value)\n\u003e       value_shape \u003d torch._C._infer_size(value.size(), self.batch_shape) if self.batch_shape else value.size()\nE       RuntimeError: expected int at position 0, but got: Tensor\n```\nAfter this change I\u0027m able to jit many more of Pyro\u0027s tests.\n\nReviewed By: ezyang\n\nDifferential Revision: D9477487\n\nPulled By: apaszke\n\nfbshipit-source-id: 5f39b29c6b8fa606ad30b02fefe2dfb618e883d6\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "b1ecc5e93316fc51143594fe3bb09e1ee37c9812",
      "old_mode": 33188,
      "old_path": "torch/distributions/categorical.py",
      "new_id": "444deffd4b27caf29aa0f1d0ac8a9338773f4439",
      "new_mode": 33188,
      "new_path": "torch/distributions/categorical.py"
    },
    {
      "type": "modify",
      "old_id": "d5f3d9c98d05802429f3474f4a011246ccabb7ed",
      "old_mode": 33188,
      "old_path": "torch/distributions/lowrank_multivariate_normal.py",
      "new_id": "3b0619868649ebabd39cf7fb79fa7f6d8b2a1e31",
      "new_mode": 33188,
      "new_path": "torch/distributions/lowrank_multivariate_normal.py"
    },
    {
      "type": "modify",
      "old_id": "72300ee4e0039c1008bea012f27479cd29082521",
      "old_mode": 33188,
      "old_path": "torch/distributions/multivariate_normal.py",
      "new_id": "51de7816cafaffaae59c278c475fad02090c616f",
      "new_mode": 33188,
      "new_path": "torch/distributions/multivariate_normal.py"
    }
  ]
}