math-pathlib 0.1.0__tar.gz

math_pathlib-0.1.0/PKG-INFO

@@ -0,0 +1,5 @@
+Metadata-Version: 2.4
+Name: math_pathlib
+Version: 0.1.0
+Summary: Educational and practical Python learning resources
+Requires-Python: >=3.13

math_pathlib-0.1.0/README.md

@@ -0,0 +1,38 @@
+# matplotlib-pyplot
+
+Educational and practical Python learning resources.
+
+## Current Contents
+
+### Logic 1
+
+- 1_RFL — Reinforcement Learning
+- 2_NLP — reserved for future content
+- 3_CV — reserved for future content
+
+### Reinforcement Learning
+
+The current release contains 8 TXT-based learning resources covering topics such as:
+
+- MDP components
+- Bandit exploration
+- TD(0) learning
+- Policy iteration
+- Value iteration
+- SARSA
+- On-policy vs Off-policy methods
+- First-Visit Monte Carlo
+
+## Usage
+
+The package provides a menu-driven interface for navigating the learning resources.
+
+## Development Status
+
+Version 0.1.0
+
+This is an initial release. NLP content will be added in a future update.
+
+## Python Requirement
+
+Python 3.13 or newer.

math_pathlib-0.1.0/math_pathlib.egg-info/PKG-INFO

@@ -0,0 +1,5 @@
+Metadata-Version: 2.4
+Name: math_pathlib
+Version: 0.1.0
+Summary: Educational and practical Python learning resources
+Requires-Python: >=3.13

math_pathlib-0.1.0/math_pathlib.egg-info/SOURCES.txt

@@ -0,0 +1,20 @@
+README.md
+pyproject.toml
+math_pathlib.egg-info/PKG-INFO
+math_pathlib.egg-info/SOURCES.txt
+math_pathlib.egg-info/dependency_links.txt
+math_pathlib.egg-info/top_level.txt
+matplotlib_pyplot/__init__.py
+matplotlib_pyplot/menu.py
+matplotlib_pyplot/logic1/__init__.py
+matplotlib_pyplot/logic1/category_menu.py
+matplotlib_pyplot/logic1/menu.py
+matplotlib_pyplot/logic1/1_RFL/P_1_Model_MDP_Components.txt
+matplotlib_pyplot/logic1/1_RFL/P_2_Compare_Bandit_Exploration.txt
+matplotlib_pyplot/logic1/1_RFL/P_3_Implement_TD0_Learning.txt
+matplotlib_pyplot/logic1/1_RFL/P_4_Plot_Policy_Iteration_Convergence.txt
+matplotlib_pyplot/logic1/1_RFL/P_5_Compare_Policy_Value_Iteration.txt
+matplotlib_pyplot/logic1/1_RFL/P_6_Implement_SARSA_Control.txt
+matplotlib_pyplot/logic1/1_RFL/P_7_Compare_On_vs_Off_Policy.txt
+matplotlib_pyplot/logic1/1_RFL/P_8_Implement_First_Visit_Monte_Carlo.txt
+matplotlib_pyplot/logic1/1_RFL/menu.py

math_pathlib-0.1.0/math_pathlib.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

math_pathlib-0.1.0/math_pathlib.egg-info/top_level.txt

@@ -0,0 +1 @@
+matplotlib_pyplot

math_pathlib-0.1.0/matplotlib_pyplot/__init__.py

@@ -0,0 +1,4 @@
+from .menu import run
+
+
+__all__ = ["run"]

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_1_Model_MDP_Components.txt

@@ -0,0 +1,11 @@
+import numpy as np, pandas as pd, matplotlib.pyplot as plt
+
+P = np.array([[0.7, 0.3], [0.4, 0.6]])
+V = np.array([1.0, 0.0])
+traj = [V @ np.linalg.matrix_power(P, t) for t in range(20)]
+df = pd.DataFrame(traj, columns=["State A", "State B"])
+
+df.plot(title="MDP State Transition Trajectory")
+plt.xlabel("Time Step")
+plt.ylabel("Probability")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_2_Compare_Bandit_Exploration.txt

@@ -0,0 +1,16 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+def bandit(mode, K=5, N=1000):
+    q_true, Q, N_a = np.random.randn(K), np.zeros(K), np.zeros(K)
+    for t in range(1, N + 1):
+        bonus = np.sqrt(2 * np.log(t) / np.maximum(N_a, 1e-9))
+        a = np.argmax(Q + bonus) if mode == "ucb" else (np.random.randint(K) if np.random.rand() < 0.1 else np.argmax(Q))
+        r = q_true[a] + np.random.randn()
+        N_a[a] += 1
+        Q[a] += (r - Q[a]) / N_a[a]
+        yield r
+
+pd.DataFrame({"UCB": bandit("ucb"), "Eps-Greedy": bandit("eps")}).expanding().mean().plot(title="Bandit Exploration")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_3_Implement_TD0_Learning.txt

@@ -0,0 +1,17 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+V, g, a, V_true = np.zeros(5), 0.9, 0.1, 0.9 ** np.arange(3, -1, -1)
+errs = []
+
+for _ in range(300):
+    s = 0
+    while s < 4:
+        sn, r = s + 1, float(s == 3)
+        V[s] += a * (r + g * V[sn] - V[s])
+        s = sn
+    errs.append(np.sqrt(np.mean((V[:4] - V_true) ** 2)))
+
+pd.Series(errs).plot(title="TD(0) RMSE Convergence", xlabel="Episode", ylabel="RMSE")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_4_Plot_Policy_Iteration_Convergence.txt

@@ -0,0 +1,16 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+S, A, g = 5, 2, 0.9
+P, R = np.random.dirichlet(np.ones(S), size=(S, A)), np.random.randn(S, A)
+pi, history = np.zeros(S, dtype=int), []
+
+for _ in range(10):
+    V = np.linalg.solve(np.eye(S) - g * P[np.arange(S), pi], R[np.arange(S), pi])
+    Q = np.column_stack([R[:, a] + g * P[:, a] @ V for a in range(A)])
+    pi = np.argmax(Q, axis=1)
+    history.append(V.mean())
+
+pd.Series(history).plot(title="Policy Iteration Value Convergence", xlabel="Iteration", ylabel="Mean V")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_5_Compare_Policy_Value_Iteration.txt

@@ -0,0 +1,23 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+S, A, g = 5, 2, 0.9
+P, R = np.random.dirichlet(np.ones(S), size=(S, A)), np.random.randn(S, A)
+
+def solve(pi):
+    V, pol = np.zeros(S), np.arange(S) % A
+    for _ in range(12):
+        Q = np.column_stack([R[:, a] + g * P[:, a] @ V for a in range(A)])
+        pol = np.argmax(Q, axis=1)
+        V = np.linalg.solve(
+            np.eye(S) - g * P[np.arange(S), pol],
+            R[np.arange(S), pol]
+        ) if pi else Q.max(axis=1)
+        yield V.mean()
+
+pd.DataFrame({
+    "Policy Iteration": solve(True),
+    "Value Iteration": solve(False)
+}).plot(title="PI vs VI Convergence")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_6_Implement_SARSA_Control.txt

@@ -0,0 +1,17 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+def sarsa(episodes=300, Q=np.zeros((5, 2))):
+    pi = lambda s: np.random.randint(2) if np.random.rand() < 0.1 else np.argmax(Q[s])
+    for _ in range(episodes):
+        s, ep_r, a = 0, 0, pi(0)
+        while s < 4:
+            sn, r = min(4, max(0, s + (1 if a == 1 else -1))), (1.0 if s == 3 and a == 1 else -0.1)
+            an = pi(sn)
+            Q[s, a] += 0.1 * (r + 0.9 * Q[sn, an] - Q[s, a])
+            s, a, ep_r = sn, an, ep_r + r
+        yield ep_r
+
+pd.Series(sarsa()).rolling(20).mean().plot(title="SARSA Control Learning Curve")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_7_Compare_On_vs_Off_Policy.txt

@@ -0,0 +1,22 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+def run(off_policy, Q=np.zeros((5, 2))):
+    pi = lambda s: np.random.randint(2) if np.random.rand() < 0.2 else np.argmax(Q[s])
+    for _ in range(300):
+        s, ep_r = 0, 0
+        while s < 4:
+            a = pi(s)
+            sn = min(4, max(0, s + (1 if a == 1 else -1)))
+            r = 1.0 if s == 3 and a == 1 else -0.1
+            target = Q[sn].max() if off_policy else Q[sn, pi(sn)]
+            Q[s, a] += 0.1 * (r + 0.9 * target - Q[s, a])
+            s, ep_r = sn, ep_r + r
+        yield ep_r
+
+pd.DataFrame({
+    "Q-Learning": run(True),
+    "SARSA": run(False)
+}).rolling(20).mean().plot(title="SARSA vs Q-Learning")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/P_8_Implement_First_Visit_Monte_Carlo.txt

@@ -0,0 +1,24 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+V, returns = np.zeros(5), [[] for _ in range(5)]
+
+for _ in range(300):
+    s, episode = 0, []
+    while s < 4:
+        sn, r = s + 1, float(s == 3)
+        episode.append((s, r))
+        s = sn
+
+    G = 0
+    for state, reward in reversed(episode):
+        G = reward + 0.9 * G
+        if not any(x[0] == state for x in episode[:episode.index((state, reward))]):
+            returns[state].append(G)
+            V[state] = np.mean(returns[state])
+
+pd.Series(V).plot(kind="bar", title="First-Visit Monte Carlo State Values")
+plt.xlabel("State")
+plt.ylabel("Value")
+plt.show()

math_pathlib-0.1.0/matplotlib_pyplot/logic1/1_RFL/menu.py

@@ -0,0 +1,60 @@
+from pathlib import Path
+
+
+def discover_files():
+    """Discover all TXT files inside the RFL directory."""
+    rfl_path = Path(__file__).resolve().parent
+
+    return sorted(
+        rfl_path.glob("*.txt"),
+        key=lambda path: path.name
+    )
+
+
+def display_file(file_path):
+    """Display the complete contents of a selected TXT file."""
+
+    print("\n========================================")
+    print(file_path.name)
+    print("========================================\n")
+
+    content = file_path.read_text(encoding="utf-8")
+
+    print(content)
+
+
+def run():
+    """Display and handle the RFL file menu."""
+
+    while True:
+        files = discover_files()
+
+        print("\n========================================")
+        print("1_RFL")
+        print()
+
+        for index, file_path in enumerate(files, start=1):
+            print(f"{index}. {file_path.stem}")
+
+        print(f"{len(files) + 1}. Back")
+        print("   E. Exit")
+
+        choice = input("\nEnter option: ").strip().lower()
+
+        if choice in {"e", "exit"}:
+            print("Exiting matplotlib_pyplot.")
+            return
+
+        if choice.isdigit():
+            option = int(choice)
+
+            if 1 <= option <= len(files):
+                display_file(files[option - 1])
+
+                input("\nPress Enter to return to 1_RFL...")
+                continue
+
+            if option == len(files) + 1:
+                return
+
+        print("Invalid option. Please try again.")

math_pathlib-0.1.0/matplotlib_pyplot/logic1/category_menu.py

@@ -0,0 +1,55 @@
+from pathlib import Path
+
+
+def run_category(category_path):
+    """Display TXT files inside a category directory."""
+
+    category_path = Path(category_path)
+
+    while True:
+        files = sorted(
+            category_path.glob("*.txt"),
+            key=lambda path: path.name
+        )
+
+        print("\n========================================")
+        print(category_path.name)
+        print()
+
+        if files:
+            for index, file_path in enumerate(files, start=1):
+                print(f"{index}. {file_path.stem}")
+        else:
+            print("No TXT files found.")
+
+        print(f"{len(files) + 1}. Back")
+        print("   E. Exit")
+
+        choice = input("\nEnter option: ").strip().lower()
+
+        if choice in {"e", "exit"}:
+            return "exit"
+
+        if choice.isdigit():
+            option = int(choice)
+
+            if 1 <= option <= len(files):
+                selected_file = files[option - 1]
+
+                print("\n========================================")
+                print(selected_file.name)
+                print("========================================\n")
+
+                print(
+                    selected_file.read_text(
+                        encoding="utf-8"
+                    )
+                )
+
+                input("\nPress Enter to return...")
+                continue
+
+            if option == len(files) + 1:
+                return "back"
+
+        print("Invalid option. Please try again.")

math_pathlib-0.1.0/matplotlib_pyplot/logic1/menu.py

@@ -0,0 +1,63 @@
+from pathlib import Path
+
+from .category_menu import run_category
+
+
+def discover_categories():
+    """Discover numbered category directories inside logic1."""
+
+    logic_path = Path(__file__).resolve().parent
+
+    categories = []
+
+    for path in logic_path.iterdir():
+        if (
+            path.is_dir()
+            and "_" in path.name
+            and path.name.split("_", 1)[0].isdigit()
+        ):
+            categories.append(path)
+
+    return sorted(
+        categories,
+        key=lambda path: int(path.name.split("_", 1)[0])
+    )
+
+
+def run():
+    """Display and handle the logic1 category menu."""
+
+    while True:
+        categories = discover_categories()
+
+        print("\n========================================")
+        print("LOGIC1")
+        print()
+
+        for index, category in enumerate(categories, start=1):
+            print(f"{index}. {category.name}")
+
+        print(f"{len(categories) + 1}. Back")
+        print("   E. Exit")
+
+        choice = input("\nEnter option: ").strip().lower()
+
+        if choice in {"e", "exit"}:
+            print("Exiting matplotlib_pyplot.")
+            return
+
+        if choice.isdigit():
+            option = int(choice)
+
+            if 1 <= option <= len(categories):
+                result = run_category(categories[option - 1])
+
+                if result == "exit":
+                    return
+
+                continue
+
+            if option == len(categories) + 1:
+                return
+
+        print("Invalid option. Please try again.")

math_pathlib-0.1.0/matplotlib_pyplot/menu.py

@@ -0,0 +1,59 @@
+from pathlib import Path
+import importlib
+
+
+def discover_logic_modules():
+    """Discover logic directories inside the main package."""
+    package_path = Path(__file__).resolve().parent
+
+    logic_modules = []
+
+    for path in package_path.iterdir():
+        if (
+            path.is_dir()
+            and path.name.startswith("logic")
+            and path.name[5:].isdigit()
+        ):
+            logic_modules.append(path.name)
+
+    return sorted(
+        logic_modules,
+        key=lambda name: int(name[5:])
+    )
+
+
+def run():
+    """Display and handle the main logic menu."""
+
+    while True:
+        logic_modules = discover_logic_modules()
+
+        print("\n========================================")
+        print("LOGIC MENU")
+        print()
+
+        for index, logic_name in enumerate(logic_modules, start=1):
+            print(f"{index}. {logic_name}")
+
+        print("   E. Exit")
+
+        choice = input("\nEnter option: ").strip().lower()
+
+        if choice in {"e", "exit"}:
+            print("Exiting matplotlib_pyplot.")
+            return
+
+        if choice.isdigit():
+            option = int(choice)
+
+            if 1 <= option <= len(logic_modules):
+                selected_logic = logic_modules[option - 1]
+
+                module = importlib.import_module(
+                    f"matplotlib_pyplot.{selected_logic}.menu"
+                )
+
+                module.run()
+                continue
+
+        print("Invalid option. Please try again.")

math_pathlib-0.1.0/pyproject.toml

@@ -0,0 +1,18 @@
+[build-system]
+requires = ["setuptools>=68"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "math_pathlib"
+version = "0.1.0"
+description = "Educational and practical Python learning resources"
+requires-python = ">=3.13"
+dependencies = []
+
+[tool.setuptools.packages.find]
+include = ["matplotlib_pyplot*"]
+
+[tool.setuptools.package-data]
+matplotlib_pyplot = [
+    "logic1/1_RFL/*.txt"
+]

math_pathlib-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+