pyro test error?

gp2d.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import pyro.contrib.gp as gp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[26.230701446533203]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = torch.tensor([[1, 5, 3], [4, 3, 7]])\n",
    "y2D = torch.tensor([[1, 2], [3, 3], [1, 4], [-1, 1]])\n",
    "kernel = gp.kernels.RBF(input_dim=3, variance=torch.tensor([1]), lengthscale=torch.tensor([3]))\n",
    "noise = torch.tensor([1e-6])\n",
    "\n",
    "gp = gp.models.GPRegression(X, y2D, kernel, noise)\n",
    "gp.optimize(num_steps=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "============================= test session starts ==============================\n",
      "platform linux -- Python 3.5.4, pytest-3.3.2, py-1.5.2, pluggy-0.6.0 -- /home/fehiepsi/miniconda3/envs/pyro/bin/python\n",
      "cachedir: .cache\n",
      "benchmark: 3.1.1 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)\n",
      "rootdir: /home/fehiepsi/pyro, inifile: setup.cfg\n",
      "plugins: xdist-1.22.1, forked-0.2, cov-2.5.1, benchmark-3.1.1, nbval-0.7\n",
      "collecting ... collected 1 item\n",
      "\n",
      "tests/contrib/gp/test_models.py::test_inference[GPRegression_y2D] FAILED [100%]\n",
      "\n",
      "=================================== FAILURES ===================================\n",
      "_______________________ test_inference[GPRegression_y2D] _______________________\n",
      "\n",
      "model_class = <class 'pyro.contrib.gp.models.gpr.GPRegression'>\n",
      "X = \n",
      " 1  5  3\n",
      " 4  3  7\n",
      "[torch.DoubleTensor of size (2,3)]\n",
      "\n",
      "y = \n",
      " 1  2\n",
      " 3  3\n",
      " 1  4\n",
      "-1  1\n",
      "[torch.DoubleTensor of size (4,2)]\n",
      "\n",
      "kernel = RBF(\n",
      ")\n",
      "likelihood = \n",
      "1.00000e-06 *\n",
      "  1.0000\n",
      "[torch.DoubleTensor of size (1,)]\n",
      "\n",
      "\n",
      "    @pytest.mark.parametrize(\"model_class, X, y, kernel, likelihood\", TEST_CASES, ids=TEST_IDS)\n",
      "    def test_inference(model_class, X, y, kernel, likelihood):\n",
      "        if model_class is SparseGPRegression or model_class is SparseVariationalGP:\n",
      "            gp = model_class(X, y, kernel, X, likelihood)\n",
      "        else:\n",
      "            gp = model_class(X, y, kernel, likelihood)\n",
      "    \n",
      ">       gp.optimize(num_steps=1)\n",
      "\n",
      "tests/contrib/gp/test_models.py:110: \n",
      "_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n",
      "pyro/contrib/gp/models/model.py:76: in optimize\n",
      "    losses.append(svi.step())\n",
      "pyro/infer/svi.py:97: in step\n",
      "    loss = self.loss_and_grads(self.model, self.guide, *args, **kwargs)\n",
      "pyro/infer/trace_elbo.py:84: in loss_and_grads\n",
      "    for model_trace, guide_trace in self._get_traces(model, guide, *args, **kwargs):\n",
      "pyro/infer/trace_elbo.py:50: in _get_traces\n",
      "    check_site_shape(site, self.max_iarange_nesting)\n",
      "_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n",
      "\n",
      "site = {'args': (), 'batch_log_pdf': \n",
      "-4.0054\n",
      "-9.3196\n",
      "-9.8386\n",
      "-3.0670\n",
      "[torch.DoubleTensor of size (4,)]\n",
      ", 'cond_indep_stack': (), 'continuation': None, ...}\n",
      "max_iarange_nesting = inf\n",
      "\n",
      "    def check_site_shape(site, max_iarange_nesting):\n",
      "        actual_shape = list(site[\"batch_log_pdf\"].shape)\n",
      "    \n",
      "        # Compute expected shape.\n",
      "        expected_shape = []\n",
      "        for f in site[\"cond_indep_stack\"]:\n",
      "            if f.dim is not None:\n",
      "                # Use the specified iarange dimension, which counts from the right.\n",
      "                assert f.dim < 0\n",
      "                if len(expected_shape) < -f.dim:\n",
      "                    expected_shape = [None] * (-f.dim - len(expected_shape)) + expected_shape\n",
      "                if expected_shape[f.dim] is not None:\n",
      "                    raise ValueError('\\n  '.join([\n",
      "                        'at site \"{}\" within iarange(\"\", dim={}), dim collision'.format(site[\"name\"], f.name, f.dim),\n",
      "                        'Try setting dim arg in other iaranges.']))\n",
      "                expected_shape[f.dim] = f.size\n",
      "        expected_shape = [1 if e is None else e for e in expected_shape]\n",
      "    \n",
      "        # Check for iarange stack overflow.\n",
      "        if len(expected_shape) > max_iarange_nesting:\n",
      "            raise ValueError('\\n  '.join([\n",
      "                'at site \"{}\", iarange stack overflow'.format(site[\"name\"]),\n",
      "                'Try increasing max_iarange_nesting to at least {}'.format(len(expected_shape))]))\n",
      "    \n",
      "        # Ignore dimensions left of max_iarange_nesting.\n",
      "        if max_iarange_nesting < len(actual_shape):\n",
      "            actual_shape = actual_shape[len(actual_shape) - max_iarange_nesting:]\n",
      "    \n",
      "        # Check for incorrect iarange placement on the right of max_iarange_nesting.\n",
      "        for actual_size, expected_size in zip_longest(reversed(actual_shape), reversed(expected_shape), fillvalue=1):\n",
      "            if expected_size != -1 and expected_size != actual_size:\n",
      "                raise ValueError('\\n  '.join([\n",
      "                    'at site \"{}\", invalid log_prob shape'.format(site[\"name\"]),\n",
      "                    'Expected {}, actual {}'.format(expected_shape, actual_shape),\n",
      "                    'Try one of the following fixes:',\n",
      "                    '- enclose the batched tensor in a with iarange(...): context',\n",
      "                    '- .reshape(extra_event_dims=...) the distribution being sampled',\n",
      ">                   '- .permute() data dimensions']))\n",
      "E               ValueError: at site \"y\", invalid log_prob shape\n",
      "E                 Expected [], actual [4]\n",
      "E                 Try one of the following fixes:\n",
      "E                 - enclose the batched tensor in a with iarange(...): context\n",
      "E                 - .reshape(extra_event_dims=...) the distribution being sampled\n",
      "E                 - .permute() data dimensions\n",
      "\n",
      "pyro/util.py:406: ValueError\n",
      "=========================== 1 failed in 0.12 seconds ===========================\n"
     ]
    }
   ],
   "source": [
    "%cat ~/pyro/error.txt"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (pyro)",
   "language": "python",
   "name": "pyro"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}