Merge branch 'release/0.8.0'

contributors_and_credits.txt

@@ -1,6 +1,7 @@
 The following people contributed to the development of this package:
 
 CERN contributors:
+ - B. Abreu Figueiredo
  - R. De Maria
  - G. Iadarola
  - S. Lopaciuk

examples/optimize/000_basic.py

@@ -0,0 +1,15 @@
+import xobjects as xo
+import xdeps as xd
+
+
+def my_function(x):
+    return [(x[0]-0.0001)**2, (x[1]-0.0003)**2, (x[2]+0.0005)**2, 3.0]
+
+x0 = [0., 0., 0.]
+
+opt = xd.Optimize.from_callable(my_function, x0=x0, steps=[1e-6, 1e-6, 1e-6, 1e-6],
+                        tar=[0., 0., 0., 3.], tols=[1e-12, 1e-12, 1e-12, 1e-12])
+opt.solve()
+
+xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+

examples/optimize/001_rescale_vars.py

@@ -0,0 +1,59 @@
+import xobjects as xo
+import xdeps as xd
+import numpy as np
+
+
+def my_function(x):
+    return x
+
+x0 = [0., 0., 0.]
+
+opt = xd.Optimize.from_callable(my_function, x0=x0, steps=[1e-6, 1e-6, 1e-6],
+                        tar=[0., 0., 0.], tols=[1e-12, 1e-12, 1e-12],
+                        limits=[[-1, 2], [-1, 4], [-2, 2]])
+opt.solve()
+
+mf = opt.get_merit_function()
+jmf = mf.get_jacobian([0.5, 2 , -1])
+xo.assert_allclose(jmf, [[1, 0, 0], [0, 1, 0], [0, 0, 1]], atol=1e-6, rtol=0)
+
+# Check scalar function
+scalar_mf = opt.get_merit_function(return_scalar=True)
+xo.assert_allclose(scalar_mf([0.5, 2, -1]), (mf([0.5, 2, -1])**2).sum(), atol=1e-10, rtol=0)
+#  Jacobian with finite differences
+jmf_scalare_ref = np.zeros(3)
+for ii in range(3):
+    dx = 1e-10
+    x1 = np.array([0.5, 2, -1])
+    x1[ii] += dx
+    jmf_scalare_ref[ii] = (scalar_mf(x1) - scalar_mf([0.5, 2, -1]))/dx
+xo.assert_allclose(jmf_scalare_ref, scalar_mf.get_jacobian([0.5, 2, -1]), atol=1e-6, rtol=0)
+
+# Check rescaling
+scaled_mf = opt.get_merit_function(rescale_x=(0, 1))
+x = [0, 1, 0.5]
+xo.assert_allclose(scaled_mf(x), [-1, 4, 0], atol=1e-10, rtol=0)
+
+# Compute jacobian of rescaled merit function using finite differences
+x0 = [0.1, 0.2, 0.3]
+jsmf_ref = np.zeros((3, 3))
+for ii in range(3):
+    dx = 1e-6
+    x1 = np.array(x0)
+    x1[ii] += dx
+    jsmf_ref[:, ii] = (scaled_mf(x1) - scaled_mf(x0))/dx
+
+jsmf = scaled_mf.get_jacobian(x0)
+xo.assert_allclose(jsmf, jsmf_ref, atol=1e-6, rtol=0)
+
+scalar_scaled_smf = opt.get_merit_function(rescale_x=(0, 1), return_scalar=True)
+xo.assert_allclose(scalar_scaled_smf(x), (scaled_mf(x)**2).sum(), atol=1e-10, rtol=0)
+
+jscarlar_scaled_smf_ref = np.zeros(3)
+for ii in range(3):
+    dx = 1e-10
+    x1 = np.array(x0)
+    x1[ii] += dx
+    jscarlar_scaled_smf_ref[ii] = (scalar_scaled_smf(x1) - scalar_scaled_smf(x0))/dx
+
+xo.assert_allclose(jscarlar_scaled_smf_ref, scalar_scaled_smf.get_jacobian(x0), atol=1e-6, rtol=0)

examples/optimize/002_scipy_algorithms.py

@@ -0,0 +1,36 @@
+import xobjects as xo
+import xdeps as xd
+import numpy as np
+
+def my_function(x):
+    return [(x[0]-0.0001)**2, (x[1]-0.0003)**2, (x[2]+0.0005)**2, 3.0]
+
+def scalar_func(x):
+    return np.sum(np.array(my_function(x))**2)
+
+x0 = [0., 0., 0.]
+limits = [[-1, 1], [-1, 1], [-1, 1]]
+targets = [0., 0., 0., 3.]
+steps = [1e-6, 1e-6, 1e-6, 1e-6]
+tols = [1e-12, 1e-12, 1e-12, 1e-12]
+
+opt = xd.Optimize.from_callable(my_function, x0=x0, steps=steps, tar=targets, tols=tols, limits=limits)
+opt.run_bfgs()
+xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+opt.reload(0)
+opt.run_l_bfgs_b()
+xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+opt.reload(0)
+opt.run_ls_trf()
+xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+opt.reload(0)
+opt.run_ls_dogbox()
+xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+assert 'bfgs' in opt.log()['tag']
+assert 'l-bfgs-b' in opt.log()['tag']
+assert 'trf' in opt.log()['tag']
+assert 'dogbox' in opt.log()['tag']

tests/test_optimize.py

@@ -0,0 +1,113 @@
+import numpy as np
+import xobjects as xo
+import xdeps as xd
+
+
+def test_mirit_function_view():
+
+    def my_function(x):
+        return x
+
+    x0 = [0., 0., 0.]
+
+    opt = xd.Optimize.from_callable(my_function, x0=x0, steps=[1e-6, 1e-6, 1e-6],
+                                    tar=[0., 0., 0.], tols=[1e-12, 1e-12, 1e-12],
+                                    limits=[[-1, 2], [-1, 4], [-2, 2]])
+    opt.solve()
+
+    mf = opt.get_merit_function()
+    jmf = mf.get_jacobian([0.5, 2, -1])
+    xo.assert_allclose(jmf, [[1, 0, 0], [0, 1, 0], [
+                       0, 0, 1]], atol=1e-6, rtol=0)
+
+    # Check scalar function
+    scalar_mf = opt.get_merit_function(return_scalar=True)
+    xo.assert_allclose(scalar_mf([0.5, 2, -1]),
+                       (mf([0.5, 2, -1])**2).sum(), atol=1e-10, rtol=0)
+    #  Jacobian with finite differences
+    jmf_scalare_ref = np.zeros(3)
+    for ii in range(3):
+        dx = 1e-10
+        x1 = np.array([0.5, 2, -1])
+        x1[ii] += dx
+        jmf_scalare_ref[ii] = (scalar_mf(x1) - scalar_mf([0.5, 2, -1]))/dx
+    xo.assert_allclose(jmf_scalare_ref, scalar_mf.get_jacobian(
+        [0.5, 2, -1]), atol=1e-6, rtol=0)
+
+    # Check rescaling
+    scaled_mf = opt.get_merit_function(rescale_x=(0, 1))
+    x = [0, 1, 0.5]
+    xo.assert_allclose(scaled_mf(x), [-1, 4, 0], atol=1e-10, rtol=0)
+
+    # Compute jacobian of rescaled merit function using finite differences
+    x0 = [0.1, 0.2, 0.3]
+    jsmf_ref = np.zeros((3, 3))
+    for ii in range(3):
+        dx = 1e-6
+        x1 = np.array(x0)
+        x1[ii] += dx
+        jsmf_ref[:, ii] = (scaled_mf(x1) - scaled_mf(x0))/dx
+
+    jsmf = scaled_mf.get_jacobian(x0)
+    xo.assert_allclose(jsmf, jsmf_ref, atol=1e-6, rtol=0)
+
+    scalar_scaled_smf = opt.get_merit_function(
+        rescale_x=(0, 1), return_scalar=True)
+    xo.assert_allclose(scalar_scaled_smf(
+        x), (scaled_mf(x)**2).sum(), atol=1e-10, rtol=0)
+
+    jscarlar_scaled_smf_ref = np.zeros(3)
+    for ii in range(3):
+        dx = 1e-10
+        x1 = np.array(x0)
+        x1[ii] += dx
+        jscarlar_scaled_smf_ref[ii] = (
+            scalar_scaled_smf(x1) - scalar_scaled_smf(x0))/dx
+
+    xo.assert_allclose(jscarlar_scaled_smf_ref,
+                       scalar_scaled_smf.get_jacobian(x0), atol=1e-6, rtol=0)
+
+def test_optimize_basics():
+
+    def my_function(x):
+        return [(x[0]-0.0001)**2, (x[1]-0.0003)**2, (x[2]+0.0005)**2, 3.0]
+
+    x0 = [0., 0., 0.]
+
+    opt = xd.Optimize.from_callable(my_function, x0=x0, steps=[1e-6, 1e-6, 1e-6, 1e-6],
+                            tar=[0., 0., 0., 3.], tols=[1e-12, 1e-12, 1e-12, 1e-12])
+    opt.solve()
+
+    xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+def test_optimize_scipy_algorithms():
+
+    def my_function(x):
+        return [(x[0]-0.0001)**2, (x[1]-0.0003)**2, (x[2]+0.0005)**2, 3.0]
+
+    x0 = [0., 0., 0.]
+    limits = [[-1, 1], [-1, 1], [-1, 1]]
+    targets = [0., 0., 0., 3.]
+    steps = [1e-6, 1e-6, 1e-6, 1e-6]
+    tols = [1e-12, 1e-12, 1e-12, 1e-12]
+
+    opt = xd.Optimize.from_callable(my_function, x0=x0, steps=steps, tar=targets, tols=tols, limits=limits)
+    opt.run_bfgs()
+    xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+    opt.reload(0)
+    opt.run_l_bfgs_b()
+    xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+    opt.reload(0)
+    opt.run_ls_trf()
+    xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+    opt.reload(0)
+    opt.run_ls_dogbox()
+    xo.assert_allclose(opt.get_merit_function().get_x(), [0.0001, 0.0003, -0.0005], atol=1e-6, rtol=0)
+
+    assert 'bfgs' in opt.log()['tag']
+    assert 'l-bfgs-b' in opt.log()['tag']
+    assert 'trf' in opt.log()['tag']
+    assert 'dogbox' in opt.log()['tag']

xdeps/madxutils.py

@@ -26,6 +26,7 @@
 
     ?power: atom
         | power "^" atom    -> pow
+        | power "**" atom    -> pow
 
     ?atom: NUMBER           -> number
          | "-" atom         -> neg
@@ -35,7 +36,7 @@
          | NAME "(" sum ("," sum)* ")" -> call
          | "(" sum ")"
 
-    NAME: /[a-z_\\.][a-z0-9_\\.%]*/
+    NAME: /[A-z_\\.][A-z0-9_\\.%]*/
     %import common.NUMBER
     %import common.WS_INLINE
     %ignore WS_INLINE