diff --git a/HW03/hw3.ipynb b/HW03/hw3.ipynb index 7c576c9..9935951 100644 --- a/HW03/hw3.ipynb +++ b/HW03/hw3.ipynb @@ -9,24 +9,122 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "metadata": {}, "outputs": [ { - "ename": "ModuleNotFoundError", - "evalue": "No module named 'numpy'", - "output_type": "error", - "traceback": [ - "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[1;31mModuleNotFoundError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[1;32mIn[1], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21;01mnumpy\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21;01mnp\u001b[39;00m\n\u001b[0;32m 2\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21;01mmatplotlib\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpyplot\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21;01mplt\u001b[39;00m\n\u001b[0;32m 4\u001b[0m \u001b[38;5;66;03m# Baseline parameters\u001b[39;00m\n", - "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'numpy'" - ] + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ + "## HW3 Problem 3\n", + "\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "\n", + "def simulate_diffusion(v, a, beta, tau, dt=1e-3, scale=1.0, max_time=10., rng=None):\n", + " \"\"\"\n", + " Simulates one realization of the diffusion process given\n", + " a set of parameters and a step size `dt`.\n", + "\n", + " Parameters:\n", + " -----------\n", + " v : float\n", + " The drift rate (rate of information uptake)\n", + " a : float\n", + " The boundary separation (decision threshold).\n", + " beta : float in [0, 1]\n", + " Relative starting point (prior option preferences)\n", + " tau : float\n", + " Non-decision time (additive constant)\n", + " dt : float, optional (default: 1e-3 = 0.001)\n", + " The step size for the Euler algorithm.\n", + " scale : float, optional (default: 1.0)\n", + " The scale (sqrt(var)) of the Wiener process. Not considered\n", + " a parameter and typically fixed to either 1.0 or 0.1.\n", + " max_time : float, optional (default: .10)\n", + " The maximum number of seconds before forced termination.\n", + " rng : np.random.Generator or None, optional (default: None)\n", + " A random number generator with locally set seed or None\n", + " If None provided, a new generator will be spawned within the function.\n", + " \n", + " Returns:\n", + " --------\n", + " (x, c) - a tuple of response time (y - float) and a \n", + " binary decision (c - int) \n", + " \"\"\"\n", + "\n", + " # Inits (process starts at relative starting point)\n", + " y = beta * a\n", + " num_steps = tau\n", + " const = scale * np.sqrt(dt)\n", + " if rng is None:\n", + " rng = np.random.default_rng()\n", + "\n", + " # Loop through process and check boundary conditions\n", + " while (y <= a and y >= 0) and num_steps <= max_time:\n", + " # Perform diffusion equation\n", + " z = rng.normal()\n", + " y += v * dt + const * z\n", + "\n", + " # Increment step counter\n", + " num_steps += dt\n", + "\n", + " if y >= a:\n", + " c = 1.\n", + " else:\n", + " c = 0.\n", + " return (round(num_steps, 4), c)\n", + "\n", + "def simulate_diffusion_n(num_sims, v, a, beta, tau, dt=1e-3, scale=1.0, max_time=10., rng=None):\n", + " \"\"\"Simulate multiple realizations of the diffusion process.\"\"\"\n", + " \n", + " # Inits\n", + " data = np.zeros((num_sims, 2))\n", + " if rng is None:\n", + " rng = np.random.default_rng()\n", + " \n", + " # Create data set\n", + " for n in range(num_sims):\n", + " data[n, :] = simulate_diffusion(v, a, beta, tau, dt, scale, max_time, rng)\n", + " return data\n", + "\n", + "def visualize_data(data, figsize=(10, 5)):\n", + " \"\"\"Helper function to visualize a simple response time data set.\"\"\"\n", + "\n", + " f, axarr = plt.subplots(1, 2, figsize=figsize)\n", + " \n", + " # Histogram of response times\n", + " sns.histplot(\n", + " data[:, 0][data[:, 1] == 1], ax=axarr[0], color='#AA0000', alpha=0.8, lw=2, label=f'Response 1')\n", + " sns.histplot(\n", + " data[:, 0][data[:, 1] == 0], ax=axarr[0], color='#0000AA', alpha=0.8, lw=2, label=f'Response 0')\n", + "\n", + " # Barplot of categorical responses\n", + " response, frequency = np.unique(data[:, 1], return_counts=True)\n", + " sns.barplot(x=response.astype(np.int32), y=frequency, ax=axarr[1], alpha=0.8, color='#00AA00')\n", + "\n", + " # Labels and embelishments\n", + " axarr[0].set_xlabel('Response time (s)', fontsize=16)\n", + " axarr[0].legend(fontsize=16)\n", + " axarr[0].set_ylabel('Count', fontsize=16)\n", + " axarr[1].set_xlabel('Response', fontsize=16)\n", + " axarr[1].set_ylabel('Frequency', fontsize=16)\n", + " for ax in axarr:\n", + " sns.despine(ax=ax)\n", + " ax.grid(alpha=0.1, color='black')\n", + "\n", + " f.suptitle('Data Summary', fontsize=18)\n", + "\n", + " f.tight_layout()\n", "\n", "# Baseline parameters\n", "a = 2.0\n", @@ -57,51 +155,11 @@ "plt.title('Effect of Drift Rate on Mean RTs')\n", "plt.show()" ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Vary boundary separation (a)\n", - "a_values = np.linspace(1.0, 3.0, 25)\n", - "mean_rt_upper = []\n", - "mean_rt_lower = []\n", - "std_rt_upper = []\n", - "std_rt_lower = []\n", - "\n", - "for a in a_values:\n", - " data = simulate_diffusion_n(num_sims, v, a, beta, tau, dt, scale, max_time)\n", - " mean_rt_upper.append(data[data[:, 1] == 1, 0].mean())\n", - " mean_rt_lower.append(data[data[:, 1] == 0, 0].mean())\n", - " std_rt_upper.append(data[data[:, 1] == 1, 0].std())\n", - " std_rt_lower.append(data[data[:, 1] == 0, 0].std())\n", - "\n", - "# Plot results\n", - "plt.figure(figsize=(12, 6))\n", - "plt.subplot(1, 2, 1)\n", - "plt.plot(a_values, mean_rt_upper, label='Upper Boundary (Correct)', color='maroon')\n", - "plt.plot(a_values, mean_rt_lower, label='Lower Boundary (Incorrect)', color='gray')\n", - "plt.xlabel('Boundary Separation (a)')\n", - "plt.ylabel('Mean Response Time (s)')\n", - "plt.legend()\n", - "\n", - "plt.subplot(1, 2, 2)\n", - "plt.plot(a_values, std_rt_upper, label='Upper Boundary (Correct)', color='maroon')\n", - "plt.plot(a_values, std_rt_lower, label='Lower Boundary (Incorrect)', color='gray')\n", - "plt.xlabel('Boundary Separation (a)')\n", - "plt.ylabel('Standard Deviation of RT (s)')\n", - "plt.legend()\n", - "\n", - "plt.suptitle('Effect of Boundary Separation on RT Distributions')\n", - "plt.show()" - ] } ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "venv", "language": "python", "name": "python3" }, @@ -115,7 +173,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.2" + "version": "3.12.3" } }, "nbformat": 4,