Adding the GPR Notebook for Hyperparameter tuning

GPR_Optimization.py

@@ -0,0 +1,181 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "marimo",
+#     "matplotlib==3.10.3",
+#     "numpy==2.2.6",
+#     "scikit-learn==1.6.1",
+#     "scipy==1.15.3",
+# ]
+# ///
+
+import marimo
+
+__generated_with = "0.13.7"
+app = marimo.App(width="medium")
+
+
+@app.cell
+def _():
+    import numpy as np
+    import matplotlib.pyplot as plt
+    from sklearn.gaussian_process import GaussianProcessRegressor
+    from sklearn.gaussian_process.kernels import Matern, WhiteKernel, ConstantKernel as C
+
+    def black_box_function(x):
+        return - (np.sin(3*x) + 0.5 * x)
+    return (
+        C,
+        GaussianProcessRegressor,
+        Matern,
+        WhiteKernel,
+        black_box_function,
+        np,
+        plt,
+    )
+
+
+@app.cell
+def _(black_box_function, np, plt):
+    X = np.linspace(0, 5.5, 1000).reshape(-1, 1)
+    y = black_box_function(X)
+    plt.plot(X, y)
+    plt.title("Black-box function")
+    plt.xlabel("x")
+    plt.ylabel("f(x)")
+    plt.show()
+    return X, y
+
+
+@app.cell
+def _(black_box_function, np):
+    X_grid = np.linspace(0, 2, 100).reshape(-1, 1)
+    y_grid = black_box_function(X_grid)
+    x_best = X_grid[np.argmax(y_grid)]
+    return
+
+
+@app.cell
+def _(black_box_function, np):
+    # Initial sample points (simulate prior evaluations)
+    X_sample = np.array([[1.0], [3.0], [5.5]])
+    y_sample = black_box_function(X_sample)
+    return X_sample, y_sample
+
+
+@app.cell
+def _(C, GaussianProcessRegressor, Matern, WhiteKernel, X_sample, y_sample):
+    # Define the kernel
+    kernel = C(1.0) * Matern(length_scale=1.0, nu=2.5) + WhiteKernel(noise_level=1e-5, noise_level_bounds=(1e-10, 1e1))
+
+    # Create and fit the Gaussian Process model
+    gpr = GaussianProcessRegressor(kernel=kernel, alpha=0.0)
+    gpr.fit(X_sample, y_sample)
+    return (gpr,)
+
+
+@app.cell
+def _(X, X_sample, gpr, plt, y, y_sample):
+    # Predict across the domain
+    mu, std = gpr.predict(X, return_std=True)
+
+    # Plot the result
+    plt.figure(figsize=(10, 5))
+    plt.plot(X, y, 'k--', label="True function")
+    plt.plot(X, mu, 'b-', label="GPR mean")
+    plt.fill_between(X.ravel(), mu - std, mu + std, alpha=0.3, label="Uncertainty")
+    plt.scatter(X_sample, y_sample, c='red', label="Samples")
+    plt.legend()
+    plt.title("Gaussian Process Fit")
+    plt.xlabel("x")
+    plt.ylabel("f(x)")
+    plt.show()
+    return
+
+
+@app.cell
+def _(np):
+    from scipy.stats import norm
+
+    def expected_improvement(X, X_sample, y_sample, model, xi=0.01):
+        mu, std = model.predict(X, return_std=True)
+        mu_sample_opt = np.min(y_sample)
+
+        with np.errstate(divide='warn'):
+            imp = mu_sample_opt - mu - xi  # because we are minimizing
+            Z = imp / std
+            ei = imp * norm.cdf(Z) + std * norm.pdf(Z)
+            ei[std == 0.0] = 0.0
+
+        return ei
+
+    return (expected_improvement,)
+
+
+@app.cell
+def _(X, X_sample, expected_improvement, gpr, np, plt, y_sample):
+    ei = expected_improvement(X, X_sample, y_sample, gpr)
+
+    plt.figure(figsize=(10, 4))
+    plt.plot(X, ei, label="Expected Improvement")
+    plt.axvline(X[np.argmax(ei)], color='r', linestyle='--', label="Next sample point")
+    plt.title("Acquisition Function (Expected Improvement)")
+    plt.xlabel("x")
+    plt.ylabel("EI(x)")
+    plt.legend()
+    plt.show()
+
+    return
+
+
+@app.cell
+def _(X, black_box_function, expected_improvement, gpr, np):
+    def bayesian_optimization(n_iter=10):
+        # Initial data
+        X_sample = np.array([[1.0], [2.5], [4.0]])
+        y_sample = black_box_function(X_sample)
+
+        for i in range(n_iter):
+            gpr.fit(X_sample, y_sample)
+            ei = expected_improvement(X, X_sample, y_sample, gpr)
+            x_next = X[np.argmax(ei)].reshape(-1, 1)
+
+            # Evaluate the function at the new point
+            y_next = black_box_function(x_next)
+
+            # Add the new sample to our dataset
+            X_sample = np.vstack((X_sample, x_next))
+            y_sample = np.append(y_sample, y_next)
+        return X_sample, y_sample
+
+    return (bayesian_optimization,)
+
+
+@app.cell
+def _(bayesian_optimization):
+    X_opt, y_opt = bayesian_optimization(n_iter=10)
+
+    return X_opt, y_opt
+
+
+@app.cell
+def _(X, X_opt, black_box_function, plt, y_opt):
+    # Plot final sampled points
+    plt.plot(X, black_box_function(X), 'k--', label="True function")
+    plt.scatter(X_opt, y_opt, c='red', label="Sampled Points")
+    plt.title("Bayesian Optimization with Gaussian Process")
+    plt.xlabel("x")
+    plt.ylabel("f(x)")
+    plt.legend()
+    plt.show()
+
+    return
+
+
+@app.cell
+def _():
+    return
+
+
+if __name__ == "__main__":
+    app.run()

public/index.html

@@ -85,22 +85,14 @@ <h2 class="notebook-title">polars vs pandas</h2>
             <h2 class="notebook-title">pyspark parametrize</h2>
             <a href="data_science_tools/pyspark_parametrize.html" class="notebook-link">View the notebook</a>
         </li>
-        <li class="notebook-item">
-            <h2 class="notebook-title">diffbot llm</h2>
-            <a href="llm/diffbot_llm.html" class="notebook-link">View the notebook</a>
-        </li>
-        <li class="notebook-item">
-            <h2 class="notebook-title">lchain deepseek</h2>
-            <a href="llm/lchain_deepseek.html" class="notebook-link">View the notebook</a>
-        </li>
-        <li class="notebook-item">
-            <h2 class="notebook-title">lchain ollama</h2>
-            <a href="llm/lchain_ollama.html" class="notebook-link">View the notebook</a>
-        </li>
         <li class="notebook-item">
             <h2 class="notebook-title">pydantic ai examples</h2>
             <a href="llm/pydantic_ai_examples.html" class="notebook-link">View the notebook</a>
         </li>
+        <li class="notebook-item">
+            <h2 class="notebook-title">temp</h2>
+            <a href="temp/temp.html" class="notebook-link">View the notebook</a>
+        </li>
     </ul>
 </body>
 </html>

public/temp.html

@@ -0,0 +1,82 @@
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8" />
+    <link rel="icon" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/favicon.ico" />
+    <!-- Preload is necessary because we show these images when we disconnect from the server,
+    but at that point we cannot load these images from the server -->
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/gradient-yHQUC_QB.png" as="image" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/noise-60BoTA8O.png" as="image" />
+    <!-- Preload the fonts -->
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/Lora-VariableFont_wght-B2ootaw-.ttf" as="font" crossorigin="anonymous" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/PTSans-Regular-CxL0S8W7.ttf" as="font" crossorigin="anonymous" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/PTSans-Bold-D9fedIX3.ttf" as="font" crossorigin="anonymous" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/FiraMono-Regular-BTCkDNvf.ttf" as="font" crossorigin="anonymous" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/FiraMono-Medium-DU3aDxX5.ttf" as="font" crossorigin="anonymous" />
+    <link rel="preload" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/FiraMono-Bold-CLVRCuM9.ttf" as="font" crossorigin="anonymous" />
+
+    <meta name="viewport" content="width=device-width, initial-scale=1" />
+    <meta name="theme-color" content="#000000" />
+    <meta name="description" content="a marimo app" />
+    <link rel="apple-touch-icon" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/apple-touch-icon.png" />
+    <link rel="manifest" crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/manifest.json" />
+
+    <script data-marimo="true">
+      function __resizeIframe(obj) {
+        var scrollbarHeight = 20; // Max between windows, mac, and linux
+
+        function setHeight() {
+          var element = obj.contentWindow.document.documentElement;
+          // If there is no vertical scrollbar, we don't need to resize the iframe
+          if (element.scrollHeight === element.clientHeight) {
+            return;
+          }
+
+          // Create a new height that includes the scrollbar height if it's visible
+          var hasHorizontalScrollbar = element.scrollWidth > element.clientWidth;
+          var newHeight = element.scrollHeight + (hasHorizontalScrollbar ? scrollbarHeight : 0);
+
+          // Only update the height if it's different from the current height
+          if (obj.style.height !== `${newHeight}px`) {
+            obj.style.height = `${newHeight}px`;
+          }
+        }
+
+        // Resize the iframe to the height of the content and bottom scrollbar height
+        setHeight();
+
+        // Resize the iframe when the content changes
+        const resizeObserver = new ResizeObserver((entries) => {
+          setHeight();
+        });
+        resizeObserver.observe(obj.contentWindow.document.body);
+      }
+    </script>
+    <marimo-filename hidden>temp.py</marimo-filename>
+    <marimo-mode data-mode='read' hidden></marimo-mode>
+    <marimo-version data-version='0.13.7' hidden></marimo-version>
+    <marimo-user-config data-config='{"completion": {"activate_on_typing": true, "copilot": false}, "display": {"cell_output": "above", "code_editor_font_size": 14, "dataframes": "rich", "default_table_page_size": 10, "default_width": "medium", "theme": "light"}, "formatting": {"line_length": 79}, "keymap": {"overrides": {}, "preset": "default"}, "language_servers": {"pylsp": {"enable_flake8": false, "enable_mypy": true, "enable_pydocstyle": false, "enable_pyflakes": false, "enable_pylint": false, "enable_ruff": true, "enabled": true}}, "package_management": {"manager": "uv"}, "runtime": {"auto_instantiate": true, "auto_reload": "off", "on_cell_change": "autorun", "output_max_bytes": 8000000, "reactive_tests": true, "std_stream_max_bytes": 1000000, "watcher_on_save": "lazy"}, "save": {"autosave": "after_delay", "autosave_delay": 1000, "format_on_save": false}, "server": {"browser": "default", "follow_symlink": false}, "snippets": {"custom_paths": [], "include_default_snippets": true}}' data-overrides='{}' hidden></marimo-user-config>
+    <marimo-app-config data-config='{"sql_output": "auto", "width": "medium"}' hidden></marimo-app-config>
+    <marimo-server-token data-token='static' hidden></marimo-server-token>
+    <title>temp</title>
+    <script type="module" crossorigin crossorigin="anonymous" src="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/index-4h4O6nRv.js"></script>
+    <link rel="stylesheet" crossorigin crossorigin="anonymous" href="https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist/assets/index-DredxwTT.css">
+
+<script data-marimo="true">
+    window.__MARIMO_STATIC__ = {};
+    window.__MARIMO_STATIC__.version = "0.13.7";
+    window.__MARIMO_STATIC__.notebookState = {"cellIds": ["Hbol", "MJUe", "vblA"], "cellNames": ["Xw==", "Xw==", "Xw=="], "cellCodes": ["", "aW1wb3J0JTIwbnVtcHklMjBhcyUyMG5wJTIwJTBBWCUyMCUzRCUyMG5wLmxpbnNwYWNlKDAlMkMxMCUyQzEwMCklMEFwcmludChYKQ==", ""], "cellConfigs": ["JTdCJTIyY29sdW1uJTIyJTNBJTIwbnVsbCUyQyUyMCUyMmRpc2FibGVkJTIyJTNBJTIwZmFsc2UlMkMlMjAlMjJoaWRlX2NvZGUlMjIlM0ElMjBmYWxzZSU3RA==", "JTdCJTIyY29sdW1uJTIyJTNBJTIwbnVsbCUyQyUyMCUyMmRpc2FibGVkJTIyJTNBJTIwZmFsc2UlMkMlMjAlMjJoaWRlX2NvZGUlMjIlM0ElMjBmYWxzZSU3RA==", "JTdCJTIyY29sdW1uJTIyJTNBJTIwbnVsbCUyQyUyMCUyMmRpc2FibGVkJTIyJTNBJTIwZmFsc2UlMkMlMjAlMjJoaWRlX2NvZGUlMjIlM0ElMjBmYWxzZSU3RA=="], "cellOutputs": {"Hbol": "JTdCJTIyY2hhbm5lbCUyMiUzQSUyMCUyMm91dHB1dCUyMiUyQyUyMCUyMm1pbWV0eXBlJTIyJTNBJTIwJTIydGV4dCUyRnBsYWluJTIyJTJDJTIwJTIyZGF0YSUyMiUzQSUyMCUyMiUyMiUyQyUyMCUyMnRpbWVzdGFtcCUyMiUzQSUyMDE3NDgyOTIyMTguNTk0NzUzJTdE", "MJUe": "JTdCJTIyY2hhbm5lbCUyMiUzQSUyMCUyMm91dHB1dCUyMiUyQyUyMCUyMm1pbWV0eXBlJTIyJTNBJTIwJTIydGV4dCUyRnBsYWluJTIyJTJDJTIwJTIyZGF0YSUyMiUzQSUyMCUyMiUyMiUyQyUyMCUyMnRpbWVzdGFtcCUyMiUzQSUyMDE3NDgyOTIyMTkuOTI2OTAzJTdE", "vblA": "JTdCJTIyY2hhbm5lbCUyMiUzQSUyMCUyMm91dHB1dCUyMiUyQyUyMCUyMm1pbWV0eXBlJTIyJTNBJTIwJTIydGV4dCUyRnBsYWluJTIyJTJDJTIwJTIyZGF0YSUyMiUzQSUyMCUyMiUyMiUyQyUyMCUyMnRpbWVzdGFtcCUyMiUzQSUyMDE3NDgyOTIyMTkuOTMwNTg0JTdE"}, "cellConsoleOutputs": {"MJUe": ["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"]}};
+    window.__MARIMO_STATIC__.assetUrl = "https://cdn.jsdelivr.net/npm/@marimo-team/[email protected]/dist";
+    window.__MARIMO_STATIC__.files = {};
+</script>
+</head>
+  <body>
+    <div id="root"></div>
+
+<marimo-code hidden="">
+    import%20marimo%0A%0A__generated_with%20%3D%20%220.13.7%22%0Aapp%20%3D%20marimo.App(width%3D%22medium%22)%0A%0A%0A%40app.cell%0Adef%20_()%3A%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_()%3A%0A%20%20%20%20import%20numpy%20as%20np%20%0A%20%20%20%20X%20%3D%20np.linspace(0%2C10%2C100)%0A%20%20%20%20print(X)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_()%3A%0A%20%20%20%20return%0A%0A%0Aif%20__name__%20%3D%3D%20%22__main__%22%3A%0A%20%20%20%20app.run()%0A
+</marimo-code>
+
+<marimo-code-hash hidden="">aedf1ff40fcab818e81332aa77d9fe138097f5fa660be8fe3efd77cd485e2686</marimo-code-hash>
+</body>
+</html>