Adding reciprocal function to rewards.tolerance.

yuvaltassa · copybara-github · commit 088b0c023565 · 2020-04-20T08:32:31.000-07:00
PiperOrigin-RevId: 307403915
Change-Id: I006691b2ad397fb6e21b8d9f98d7b60e1855c4e4
diff --git a/dm_control/utils/rewards.py b/dm_control/utils/rewards.py
@@ -62,6 +62,10 @@ def _sigmoids(x, value_at_1, sigmoid):
     scale = np.sqrt(1/value_at_1 - 1)
     return 1 / ((x*scale)**2 + 1)
 
+  elif sigmoid == 'reciprocal':
+    scale = 1/value_at_1 - 1
+    return 1 / (abs(x)*scale + 1)
+
   elif sigmoid == 'cosine':
     scale = np.arccos(2*value_at_1 - 1) / np.pi
     scaled_x = x*scale
diff --git a/dm_control/utils/rewards_test.py b/dm_control/utils/rewards_test.py
@@ -45,7 +45,7 @@ def test_tolerance_sigmoid_parameterisation(self, margin, value_at_margin):
 
   @parameterized.parameters(("gaussian",), ("hyperbolic",), ("long_tail",),
                             ("cosine",), ("tanh_squared",), ("linear",),
-                            ("quadratic"))
+                            ("quadratic",), ("reciprocal",))
   def test_tolerance_sigmoids(self, sigmoid):
     margins = [0.01, 1.0, 100, 10000]
     values_at_margin = [0.1, 0.5, 0.9]
diff --git a/setup.py b/setup.py
@@ -166,7 +166,7 @@ def find_data_files(package_dir, patterns):
 
 setup(
     name='dm_control',
-    version='0.0.307384858',
+    version='0.0.307403915',
     description='Continuous control environments and MuJoCo Python bindings.',
     author='DeepMind',
     license='Apache License, Version 2.0',
