Save new bench

limix · Apr 24, 2017 · 277f573 · 277f573
1 parent 4ea8316
commit 277f573
Show file tree

Hide file tree

Showing 3 changed files with 68 additions and 50 deletions.
diff --git a/.asv/results/benchmarks.json b/.asv/results/benchmarks.json
@@ -13,33 +13,5 @@
         "unit": "seconds",
         "warmup_time": -1
     },
-    "benchmarks.TimeSuite.time_qep_binomial_lml_learn": {
-        "code": "def time_qep_binomial_lml_learn(self):\n    glmm = GLMM((self._nsuc, self._ntri), 'binomial', self._X, self._QS)\n\n    assert_allclose(glmm.value(), -272.1213895386019)\n    glmm.fix('beta')\n    glmm.fix('scale')\n\n    glmm.feed().maximize(progress=False)\n    assert_allclose(glmm.value(), -271.367864630782)\n\n    glmm.unfix('beta')\n    glmm.unfix('scale')\n\n    glmm.feed().maximize(progress=False)\n    assert_allclose(glmm.value(), -266.9517518211878)\n",
-        "goal_time": 0.1,
-        "name": "benchmarks.TimeSuite.time_qep_binomial_lml_learn",
-        "number": 0,
-        "param_names": [],
-        "params": [],
-        "pretty_name": "benchmarks.TimeSuite.time_qep_binomial_lml_learn",
-        "repeat": 0,
-        "timeout": 60.0,
-        "type": "time",
-        "unit": "seconds",
-        "warmup_time": -1
-    },
-    "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn": {
-        "code": "def time_qep_binomial_lml_no_learn(self):\n    glmm = GLMM((self._nsuc, self._ntri), 'binomial', self._X, self._QS)\n    assert_allclose(glmm.value(), -272.1213895386019)\n",
-        "goal_time": 0.1,
-        "name": "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn",
-        "number": 0,
-        "param_names": [],
-        "params": [],
-        "pretty_name": "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn",
-        "repeat": 0,
-        "timeout": 60.0,
-        "type": "time",
-        "unit": "seconds",
-        "warmup_time": -1
-    },
     "version": 1
 }
diff --git a/.asv/results/reborn/4ea83161-conda-py3.6-numba-numpy-numpy-sugar-pip-pip+optimix-scipy.json b/.asv/results/reborn/4ea83161-conda-py3.6-numba-numpy-numpy-sugar-pip-pip+optimix-scipy.json
@@ -0,0 +1,65 @@
+{
+    "commit_hash": "4ea831612c70c8b5b04bef8ab4945558f903dfea",
+    "date": 1493047105000,
+    "ended_at": {
+        "benchmarks.TimeSuite.time_qep_binomial_1k_learn": 1493047579000,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_learn": 1493047300000,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn": 1493047456000
+    },
+    "env_name": "conda-py3.6-numba-numpy-numpy-sugar-pip-pip+optimix-scipy",
+    "params": {
+        "arch": "x86_64",
+        "cpu": "Intel(R) Core(TM) i7-5557U CPU @ 3.10GHz",
+        "machine": "reborn",
+        "numba": "",
+        "numpy": "",
+        "numpy-sugar": "",
+        "os": "Darwin 16.5.0",
+        "pip": "",
+        "pip+optimix": "",
+        "python": "3.6",
+        "ram": "17179869184",
+        "scipy": ""
+    },
+    "profiles": {},
+    "python": "3.6",
+    "requirements": {
+        "numba": "",
+        "numpy": "",
+        "numpy-sugar": "",
+        "pip": "",
+        "pip+optimix": "",
+        "scipy": ""
+    },
+    "results": {
+        "benchmarks.TimeSuite.time_qep_binomial_1k_learn": null,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_learn": null,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn": {
+            "result": [
+                0.23217700000000008
+            ],
+            "stats": [
+                {
+                    "ci_99": [
+                        0.23150999999999966,
+                        0.23872900000000108
+                    ],
+                    "max": 0.23872900000000108,
+                    "mean": 0.23413866666666694,
+                    "min": 0.23150999999999966,
+                    "n": 3,
+                    "q_25": 0.23184349999999987,
+                    "q_75": 0.23545300000000058,
+                    "std": 0.0032572577764472402,
+                    "systematic": 0.0
+                }
+            ]
+        }
+    },
+    "started_at": {
+        "benchmarks.TimeSuite.time_qep_binomial_1k_learn": 1493047519000,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_learn": 1493047294000,
+        "benchmarks.TimeSuite.time_qep_binomial_lml_no_learn": 1493047452000
+    },
+    "version": 1
+}
diff --git a/benchmarks/benchmarks.py b/benchmarks/benchmarks.py
@@ -1,7 +1,7 @@
 from numpy.random import RandomState
 from numpy.testing import assert_allclose
 from numpy import ascontiguousarray, sqrt, ones
-from numpy_sugar.linalg import economic_qs, economic_qs_linear
+from numpy_sugar.linalg import economic_qs, economic_qs_linear, sum2diag
 
 from glimix_core.example import linear_eye_cov
 from glimix_core.glmm import GLMM
@@ -20,32 +20,13 @@ def setup(self):
         self._nsuc = [random.randint(0, i) for i in self._ntri]
 
         self._X1k = random.randn(n1k, 5)
-        self._K1k = linear_eye_cov().feed().value()
+        self._K1k = self._X1k.dot(self._X1k.T)
+        sum2diag(self._K1k, 1e-3, out=self._K1k)
         self._QS1k = economic_qs(self._K1k)
 
         self._ntri1k = random.randint(1, 30, n1k)
         self._nsuc1k = [random.randint(0, i) for i in self._ntri1k]
 
-    def time_qep_binomial_lml_no_learn(self):
-        glmm = GLMM((self._nsuc, self._ntri), 'binomial', self._X, self._QS)
-        assert_allclose(glmm.value(), -272.1213895386019)
-
-    def time_qep_binomial_lml_learn(self):
-        glmm = GLMM((self._nsuc, self._ntri), 'binomial', self._X, self._QS)
-
-        assert_allclose(glmm.value(), -272.1213895386019)
-        glmm.fix('beta')
-        glmm.fix('scale')
-
-        glmm.feed().maximize(progress=False)
-        assert_allclose(glmm.value(), -271.367864630782)
-
-        glmm.unfix('beta')
-        glmm.unfix('scale')
-
-        glmm.feed().maximize(progress=False)
-        assert_allclose(glmm.value(), -266.9517518211878)
-
     def time_qep_binomial_1k_learn(self):
         glmm = GLMM((self._nsuc1k, self._ntri1k), 'binomial', self._X1k,
                     self._QS1k)