mdpsolver 0.9.10__tar.gz → 0.10.0__tar.gz

{mdpsolver-0.9.10/src/mdpsolver.egg-info → mdpsolver-0.10.0}/PKG-INFO

@@ -1,19 +1,20 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.2
 Name: mdpsolver
-Version: 0.9.10
+Version: 0.10.0
 Summary: A fast solver for Markov Decision Processes
-Author-email: Anders Reenberg Andersen <andersra@live.dk>, Jesper Fink Andersen <jesperfinkandersen@gmail.com>
-Project-URL: Homepage, https://github.com/areenberg/MDPSolver
-Project-URL: Issues, https://github.com/areenberg/MDPSolver/issues
+Author-Email: Anders Reenberg Andersen <andersra@live.dk>, Jesper Fink Andersen <jesperfinkandersen@gmail.com>
 Classifier: Development Status :: 4 - Beta
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: Microsoft :: Windows
 Classifier: Operating System :: POSIX :: Linux
+Classifier: Operating System :: MacOS
+Project-URL: Homepage, https://github.com/areenberg/MDPSolver
+Project-URL: Issues, https://github.com/areenberg/MDPSolver/issues
 Requires-Python: >=3.10
+Provides-Extra: test
+Requires-Dist: numpy; extra == "test"
 Description-Content-Type: text/markdown
-License-File: LICENSE
-Dynamic: license-file
 
 # MDPSolver
 
@@ -21,7 +22,7 @@ MDPSolver is a Python package for large Markov Decision Processes (MDPs) with in
 
 ## Features
 
-* Fast solver: Our C++-based solver is substantially faster than other MDP packages available for Python. See details in the documentation.
+* Fast solver: Our C++-based solver is substantially faster than other CPU-based MDP packages available for Python. See details in the documentation.
 * Two optimality criteria: *Discounted* and *Average* reward.
 * Three optimization algorithms: *Value iteration*, *Policy iteration*, and *Modified policy iteration*.
 * Three value-update methods: *Standard*, *Gauss–Seidel*, and *Successive over-relaxation*.
@@ -30,23 +31,7 @@ MDPSolver is a Python package for large Markov Decision Processes (MDPs) with in
 
 # Installation
 
-## Linux
-
-Install directly from PyPI with:
-
-```
-pip install mdpsolver
-```
-
-### GLIBC not found
-
-Some users will encounter the `version 'GLIBC_2.38' not found` error when attempting to import MDPSolver in Python. The issue can be resolved by manually compiling and replacing the SO-file for the optimization module in the MDPSolver package. See the steps on how to solve the issue in the [documentation](https://github.com/areenberg/MDPSolver/wiki#how-to-install).
-
-## Windows
-
-Requires [Visual Studio 2022](https://visualstudio.microsoft.com/downloads/) (17.9) with MSVC C++ compiler and libraries installed.
-
-After installing Visual Studio (incl. MSVC C++ compiler and libraries), install directly from PyPI with:
+Prebuilt wheels are available for **Linux**, **Windows**, and **macOS (14+)** across Python 3.10–3.14. Install directly from PyPI with:
 
 ```
 pip install mdpsolver

{mdpsolver-0.9.10 → mdpsolver-0.10.0}/README.md

@@ -4,7 +4,7 @@ MDPSolver is a Python package for large Markov Decision Processes (MDPs) with in
 
 ## Features
 
-* Fast solver: Our C++-based solver is substantially faster than other MDP packages available for Python. See details in the documentation.
+* Fast solver: Our C++-based solver is substantially faster than other CPU-based MDP packages available for Python. See details in the documentation.
 * Two optimality criteria: *Discounted* and *Average* reward.
 * Three optimization algorithms: *Value iteration*, *Policy iteration*, and *Modified policy iteration*.
 * Three value-update methods: *Standard*, *Gauss–Seidel*, and *Successive over-relaxation*.
@@ -13,23 +13,7 @@ MDPSolver is a Python package for large Markov Decision Processes (MDPs) with in
 
 # Installation
 
-## Linux
-
-Install directly from PyPI with:
-
-```
-pip install mdpsolver
-```
-
-### GLIBC not found
-
-Some users will encounter the `version 'GLIBC_2.38' not found` error when attempting to import MDPSolver in Python. The issue can be resolved by manually compiling and replacing the SO-file for the optimization module in the MDPSolver package. See the steps on how to solve the issue in the [documentation](https://github.com/areenberg/MDPSolver/wiki#how-to-install).
-
-## Windows
-
-Requires [Visual Studio 2022](https://visualstudio.microsoft.com/downloads/) (17.9) with MSVC C++ compiler and libraries installed.
-
-After installing Visual Studio (incl. MSVC C++ compiler and libraries), install directly from PyPI with:
+Prebuilt wheels are available for **Linux**, **Windows**, and **macOS (14+)** across Python 3.10–3.14. Install directly from PyPI with:
 
 ```
 pip install mdpsolver

mdpsolver-0.10.0/pyproject.toml

@@ -0,0 +1,40 @@
+[build-system]
+requires = ["scikit-build-core>=0.10", "pybind11>=2.13"]
+build-backend = "scikit_build_core.build"
+
+[tool.scikit-build]
+# The CMake project lives outside Python/, next to the rest of the C++ sources.
+cmake.source-dir = "../CPP_Source_Code"
+build-dir = "build/{wheel_tag}"
+# CMakeLists.txt installs the extension to DESTINATION mdpsolver itself,
+# so wheel.install-dir stays at the default (wheel root) here.
+sdist.include = ["src/mdpsolver/*.py"]
+
+[tool.scikit-build.cmake.define]
+CMAKE_BUILD_TYPE = "Release"
+
+[project]
+name = "mdpsolver"
+version = "0.10.0"
+authors = [
+  { name="Anders Reenberg Andersen", email="andersra@live.dk" },
+  { name="Jesper Fink Andersen", email="jesperfinkandersen@gmail.com" },
+]
+description = "A fast solver for Markov Decision Processes"
+readme = "README.md"
+requires-python = ">=3.10"
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Programming Language :: Python :: 3",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: Microsoft :: Windows",
+    "Operating System :: POSIX :: Linux",
+    "Operating System :: MacOS"
+]
+
+[project.optional-dependencies]
+test = ["numpy"]
+
+[project.urls]
+Homepage = "https://github.com/areenberg/MDPSolver"
+Issues = "https://github.com/areenberg/MDPSolver/issues"

mdpsolver-0.10.0/tests/example_model.py

@@ -0,0 +1,139 @@
+import numpy as np
+import os
+
+# Create transition probability matrix and rewards for the 'mdpsolver' examples.
+
+
+# -----------------------------
+#   FUNCTIONS
+# -----------------------------
+
+
+def exitDist(startK, arrRate, serRate, tol=1e-9):
+    # calculate distribution over states observed by
+    # the next arrival assuming the system currently
+    # contains 'startK' customers.
+
+    g = [i * serRate / (i * serRate + arrRate) for i in range(startK, -1, -1)]
+    d = [arrRate / (i * serRate + arrRate) for i in range(startK, -1, -1)]
+
+    Dmat = np.diag(d)
+    Zmat = np.zeros((len(g), len(d)))
+    Imat = np.eye(len(g))
+    Gmat = np.zeros((len(g), len(g)))
+    for i in range(startK):
+        Gmat[i, i + 1] = g[i]
+    mat = np.block([[Gmat, Dmat], [Zmat, Imat]])
+    p = np.zeros((1, (startK + 1) * 2))
+    p[0, 0] = 1
+
+    while np.sum(p[0, : startK + 1]) >= tol:
+        p = np.matmul(p, mat)
+
+    return p[0, startK + 1 :]
+
+
+def tranMat(arrRate1, arrRate2, serRate, capacity, fileName):
+    # calculate transition probabilities and
+    # and save them in a file
+
+    ns = capacity + 1
+    arrRate = arrRate1 + arrRate2
+    pc1 = arrRate1 / arrRate
+    pc2 = 1 - pc1
+
+    if os.path.exists(fileName):
+        os.remove(fileName)
+    with open(fileName, "a") as file:
+        file.write(f"stateFrom,action,stateTo,probability\n")
+        for sidx in range(ns):  # stateFrom (customer type 1)
+            for aidx in range(2):
+                if aidx == 0:  # reject
+                    if sidx > 0:
+                        dist = exitDist(sidx, arrRate, serRate)
+                    else:
+                        dist = np.array([1])
+                    dist1 = pc1 * dist
+                    dist2 = pc2 * dist
+                    for jidx in range(sidx + 1):
+                        out = [sidx, aidx, jidx, dist1[sidx - jidx]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                    for jidx in range(ns, ns + sidx + 1):
+                        out = [sidx, aidx, jidx, dist2[sidx - (jidx - ns)]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                elif aidx == 1 and sidx < capacity:  # accept
+                    startK = sidx + 1
+                    dist = exitDist(startK, arrRate, serRate)
+                    dist1 = pc1 * dist
+                    dist2 = pc2 * dist
+                    for jidx in range(startK + 1):
+                        out = [sidx, aidx, jidx, dist1[startK - jidx]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                    for jidx in range(ns, ns + startK + 1):
+                        out = [sidx, aidx, jidx, dist2[startK - (jidx - ns)]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+
+        for sidx in range(ns):  # stateFrom (customer type 2)
+            for aidx in range(2):
+                if aidx == 0:  # reject
+                    if sidx > 0:
+                        dist = exitDist(sidx, arrRate, serRate)
+                    else:
+                        dist = np.array([1])
+                    dist1 = pc1 * dist
+                    dist2 = pc2 * dist
+                    for jidx in range(sidx + 1):
+                        out = [sidx + ns, aidx, jidx, dist1[sidx - jidx]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                    for jidx in range(ns, ns + sidx + 1):
+                        out = [sidx + ns, aidx, jidx, dist2[sidx - (jidx - ns)]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                elif aidx == 1 and sidx < capacity:  # accept
+                    startK = sidx + 1
+                    dist = exitDist(startK, arrRate, serRate)
+                    dist1 = pc1 * dist
+                    dist2 = pc2 * dist
+                    for jidx in range(startK + 1):
+                        out = [sidx + ns, aidx, jidx, dist1[startK - jidx]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+                    for jidx in range(ns, ns + startK + 1):
+                        out = [sidx + ns, aidx, jidx, dist2[startK - (jidx - ns)]]
+                        str_out = ",".join(str(num) for num in out)
+                        file.write(str_out + "\n")
+
+
+def rewards(capacity, rewardCust1, rewardCust2, fileName):
+
+    ns = capacity + 1
+
+    if os.path.exists(fileName):
+        os.remove(fileName)
+    with open(fileName, "a") as file:
+        file.write(f"stateFrom,action,reward\n")
+        for sidx in range(ns):  # stateFrom (customer type 1)
+            for aidx in range(2):
+                if aidx == 0:
+                    out = [sidx, aidx, 0]
+                    str_out = ",".join(str(num) for num in out)
+                    file.write(str_out + "\n")
+                elif aidx == 1 and sidx < capacity:
+                    out = [sidx, aidx, rewardCust1]
+                    str_out = ",".join(str(num) for num in out)
+                    file.write(str_out + "\n")
+        for sidx in range(ns):  # stateFrom (customer type 2)
+            for aidx in range(2):
+                if aidx == 0:
+                    out = [sidx + ns, aidx, 0]
+                    str_out = ",".join(str(num) for num in out)
+                    file.write(str_out + "\n")
+                elif aidx == 1 and sidx < capacity:
+                    out = [sidx + ns, aidx, rewardCust2]
+                    str_out = ",".join(str(num) for num in out)
+                    file.write(str_out + "\n")

mdpsolver-0.10.0/tests/rewards_example2.csv

@@ -0,0 +1,403 @@
+stateFrom,action,reward
+0,0,0
+0,1,1
+1,0,0
+1,1,1
+2,0,0
+2,1,1
+3,0,0
+3,1,1
+4,0,0
+4,1,1
+5,0,0
+5,1,1
+6,0,0
+6,1,1
+7,0,0
+7,1,1
+8,0,0
+8,1,1
+9,0,0
+9,1,1
+10,0,0
+10,1,1
+11,0,0
+11,1,1
+12,0,0
+12,1,1
+13,0,0
+13,1,1
+14,0,0
+14,1,1
+15,0,0
+15,1,1
+16,0,0
+16,1,1
+17,0,0
+17,1,1
+18,0,0
+18,1,1
+19,0,0
+19,1,1
+20,0,0
+20,1,1
+21,0,0
+21,1,1
+22,0,0
+22,1,1
+23,0,0
+23,1,1
+24,0,0
+24,1,1
+25,0,0
+25,1,1
+26,0,0
+26,1,1
+27,0,0
+27,1,1
+28,0,0
+28,1,1
+29,0,0
+29,1,1
+30,0,0
+30,1,1
+31,0,0
+31,1,1
+32,0,0
+32,1,1
+33,0,0
+33,1,1
+34,0,0
+34,1,1
+35,0,0
+35,1,1
+36,0,0
+36,1,1
+37,0,0
+37,1,1
+38,0,0
+38,1,1
+39,0,0
+39,1,1
+40,0,0
+40,1,1
+41,0,0
+41,1,1
+42,0,0
+42,1,1
+43,0,0
+43,1,1
+44,0,0
+44,1,1
+45,0,0
+45,1,1
+46,0,0
+46,1,1
+47,0,0
+47,1,1
+48,0,0
+48,1,1
+49,0,0
+49,1,1
+50,0,0
+50,1,1
+51,0,0
+51,1,1
+52,0,0
+52,1,1
+53,0,0
+53,1,1
+54,0,0
+54,1,1
+55,0,0
+55,1,1
+56,0,0
+56,1,1
+57,0,0
+57,1,1
+58,0,0
+58,1,1
+59,0,0
+59,1,1
+60,0,0
+60,1,1
+61,0,0
+61,1,1
+62,0,0
+62,1,1
+63,0,0
+63,1,1
+64,0,0
+64,1,1
+65,0,0
+65,1,1
+66,0,0
+66,1,1
+67,0,0
+67,1,1
+68,0,0
+68,1,1
+69,0,0
+69,1,1
+70,0,0
+70,1,1
+71,0,0
+71,1,1
+72,0,0
+72,1,1
+73,0,0
+73,1,1
+74,0,0
+74,1,1
+75,0,0
+75,1,1
+76,0,0
+76,1,1
+77,0,0
+77,1,1
+78,0,0
+78,1,1
+79,0,0
+79,1,1
+80,0,0
+80,1,1
+81,0,0
+81,1,1
+82,0,0
+82,1,1
+83,0,0
+83,1,1
+84,0,0
+84,1,1
+85,0,0
+85,1,1
+86,0,0
+86,1,1
+87,0,0
+87,1,1
+88,0,0
+88,1,1
+89,0,0
+89,1,1
+90,0,0
+90,1,1
+91,0,0
+91,1,1
+92,0,0
+92,1,1
+93,0,0
+93,1,1
+94,0,0
+94,1,1
+95,0,0
+95,1,1
+96,0,0
+96,1,1
+97,0,0
+97,1,1
+98,0,0
+98,1,1
+99,0,0
+99,1,1
+100,0,0
+101,0,0
+101,1,1000
+102,0,0
+102,1,1000
+103,0,0
+103,1,1000
+104,0,0
+104,1,1000
+105,0,0
+105,1,1000
+106,0,0
+106,1,1000
+107,0,0
+107,1,1000
+108,0,0
+108,1,1000
+109,0,0
+109,1,1000
+110,0,0
+110,1,1000
+111,0,0
+111,1,1000
+112,0,0
+112,1,1000
+113,0,0
+113,1,1000
+114,0,0
+114,1,1000
+115,0,0
+115,1,1000
+116,0,0
+116,1,1000
+117,0,0
+117,1,1000
+118,0,0
+118,1,1000
+119,0,0
+119,1,1000
+120,0,0
+120,1,1000
+121,0,0
+121,1,1000
+122,0,0
+122,1,1000
+123,0,0
+123,1,1000
+124,0,0
+124,1,1000
+125,0,0
+125,1,1000
+126,0,0
+126,1,1000
+127,0,0
+127,1,1000
+128,0,0
+128,1,1000
+129,0,0
+129,1,1000
+130,0,0
+130,1,1000
+131,0,0
+131,1,1000
+132,0,0
+132,1,1000
+133,0,0
+133,1,1000
+134,0,0
+134,1,1000
+135,0,0
+135,1,1000
+136,0,0
+136,1,1000
+137,0,0
+137,1,1000
+138,0,0
+138,1,1000
+139,0,0
+139,1,1000
+140,0,0
+140,1,1000
+141,0,0
+141,1,1000
+142,0,0
+142,1,1000
+143,0,0
+143,1,1000
+144,0,0
+144,1,1000
+145,0,0
+145,1,1000
+146,0,0
+146,1,1000
+147,0,0
+147,1,1000
+148,0,0
+148,1,1000
+149,0,0
+149,1,1000
+150,0,0
+150,1,1000
+151,0,0
+151,1,1000
+152,0,0
+152,1,1000
+153,0,0
+153,1,1000
+154,0,0
+154,1,1000
+155,0,0
+155,1,1000
+156,0,0
+156,1,1000
+157,0,0
+157,1,1000
+158,0,0
+158,1,1000
+159,0,0
+159,1,1000
+160,0,0
+160,1,1000
+161,0,0
+161,1,1000
+162,0,0
+162,1,1000
+163,0,0
+163,1,1000
+164,0,0
+164,1,1000
+165,0,0
+165,1,1000
+166,0,0
+166,1,1000
+167,0,0
+167,1,1000
+168,0,0
+168,1,1000
+169,0,0
+169,1,1000
+170,0,0
+170,1,1000
+171,0,0
+171,1,1000
+172,0,0
+172,1,1000
+173,0,0
+173,1,1000
+174,0,0
+174,1,1000
+175,0,0
+175,1,1000
+176,0,0
+176,1,1000
+177,0,0
+177,1,1000
+178,0,0
+178,1,1000
+179,0,0
+179,1,1000
+180,0,0
+180,1,1000
+181,0,0
+181,1,1000
+182,0,0
+182,1,1000
+183,0,0
+183,1,1000
+184,0,0
+184,1,1000
+185,0,0
+185,1,1000
+186,0,0
+186,1,1000
+187,0,0
+187,1,1000
+188,0,0
+188,1,1000
+189,0,0
+189,1,1000
+190,0,0
+190,1,1000
+191,0,0
+191,1,1000
+192,0,0
+192,1,1000
+193,0,0
+193,1,1000
+194,0,0
+194,1,1000
+195,0,0
+195,1,1000
+196,0,0
+196,1,1000
+197,0,0
+197,1,1000
+198,0,0
+198,1,1000
+199,0,0
+199,1,1000
+200,0,0
+200,1,1000
+201,0,0

{mdpsolver-0.9.10 → mdpsolver-0.10.0}/tests/test2.py

@@ -51,8 +51,9 @@ mdl1.mdp(
 )
 
 mdl1.solve(algorithm="vi")
-if not np.round(np.mean(mdl1.getValueVector()), 0) == 17075.0:
-    sys.exit("Model 1 failed!")
+mean1 = np.mean(mdl1.getValueVector())
+if not np.round(mean1, 0) == 17075.0:
+    sys.exit(f"Model 1 failed! mean={mean1!r} rounded={np.round(mean1, 0)!r}")
 
 # ---------------------------------------
 # POLICY ITERATION
@@ -68,8 +69,9 @@ mdl2.mdp(
 
 mdl2.solve(algorithm="pi")
 
-if not np.round(np.mean(mdl2.getValueVector()), 0) == 17075.0:
-    sys.exit("Model 2 failed!")
+mean2 = np.mean(mdl2.getValueVector())
+if not np.round(mean2, 0) == 17075.0:
+    sys.exit(f"Model 2 failed! mean={mean2!r} rounded={np.round(mean2, 0)!r}")
 
 # ---------------------------------------
 # MODIFIED POLICY ITERATION
@@ -85,7 +87,8 @@ mdl3.mdp(
 
 mdl3.solve(algorithm="mpi")
 
-if not np.round(np.mean(mdl3.getValueVector()), 0) == 17075.0:
-    sys.exit("Model 3 failed!")
+mean3 = np.mean(mdl3.getValueVector())
+if not np.round(mean3, 0) == 17075.0:
+    sys.exit(f"Model 3 failed! mean={mean3!r} rounded={np.round(mean3, 0)!r}")
 
 print("Test 2 succesfully reproduced output!")