diff --git a/bel_NN_dynamic.ipynb b/bel_NN_dynamic.ipynb index 6a2066a..9245661 100644 --- a/bel_NN_dynamic.ipynb +++ b/bel_NN_dynamic.ipynb @@ -2,24 +2,55 @@ "cells": [ { "cell_type": "code", - "execution_count": 11, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ + "import os\n", + "\n", "import numpy as np\n", "import pandas as pd\n", + "import cupy as cp\n", + "\n", "from sklearn.model_selection import train_test_split\n", "from itertools import product" ] }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPU is available. Using CuPy for GPU acceleration.\n" + ] + } + ], + "source": [ + "try:\n", + " import cupy as cp\n", + " if cp.cuda.is_available():\n", + " print(\"GPU is available. Using CuPy for GPU acceleration.\")\n", + " xp = cp\n", + " else:\n", + " print(\"GPU is not available. Falling back to NumPy on CPU.\")\n", + " xp = np\n", + "except ImportError:\n", + " print(\"CuPy not found. Using NumPy on CPU.\")\n", + " xp = np" + ] + }, + { + "cell_type": "code", + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "data = pd.read_csv('data/bel_data_test.csv')\n", - "data = np.array(data)\n", + "data = xp.array(data)\n", "\n", "# Split data\n", "X = data[:, 1:].T\n", @@ -40,9 +71,18 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Input layer size: 1024\n", + "Output layer size: 61\n" + ] + } + ], "source": [ "# Determine input and output layer sizes\n", "input_size = X_train.shape[0]\n", @@ -54,7 +94,18 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "X_train, Y_train = xp.array(X_train), xp.array(Y_train)\n", + "X_val, Y_val = xp.array(X_val), xp.array(Y_val)\n", + "X_test, Y_test = xp.array(X_test), xp.array(Y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, "metadata": {}, "outputs": [], "source": [ @@ -63,23 +114,23 @@ " L = len(layer_dims)\n", " \n", " for l in range(1, L):\n", - " params[f'W{l}'] = np.random.randn(layer_dims[l], layer_dims[l-1]) * np.sqrt(2. / layer_dims[l-1])\n", - " params[f'b{l}'] = np.zeros((layer_dims[l], 1))\n", + " params[f'W{l}'] = xp.random.randn(layer_dims[l], layer_dims[l-1]) * xp.sqrt(2. / layer_dims[l-1])\n", + " params[f'b{l}'] = xp.zeros((layer_dims[l], 1))\n", " \n", " return params" ] }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "def ReLU(Z):\n", - " return np.maximum(Z, 0)\n", + " return xp.maximum(Z, 0)\n", "\n", "def softmax(Z):\n", - " A = np.exp(Z) / sum(np.exp(Z))\n", + " A = xp.exp(Z) / sum(xp.exp(Z))\n", " return A\n", "\n", "def forward_prop(X, params):\n", @@ -91,13 +142,13 @@ " A_prev = A\n", " W = params[f'W{l}']\n", " b = params[f'b{l}']\n", - " Z = np.dot(W, A_prev) + b\n", + " Z = xp.dot(W, A_prev) + b\n", " A = ReLU(Z)\n", " caches.append((A_prev, W, b, Z))\n", "\n", " WL = params[f'W{L}']\n", " bL = params[f'b{L}']\n", - " ZL = np.dot(WL, A) + bL\n", + " ZL = xp.dot(WL, A) + bL\n", " AL = softmax(ZL)\n", " caches.append((A, WL, bL, ZL))\n", "\n", @@ -107,10 +158,14 @@ " return Z > 0\n", "\n", "def one_hot(Y):\n", - " one_hot_Y = np.zeros((Y.size, Y.max() + 1))\n", - " one_hot_Y[np.arange(Y.size), Y] = 1\n", - " one_hot_Y = one_hot_Y.T\n", - " return one_hot_Y\n", + " # one_hot_Y = xp.zeros((Y.size, Y.max() + 1))\n", + " # one_hot_Y[xp.arange(Y.size), Y] = 1\n", + " # one_hot_Y = one_hot_Y.T\n", + " # return one_hot_Y\n", + " Y = Y.astype(int)\n", + " one_hot_Y = xp.zeros((Y.size, int(xp.max(Y)) + 1))\n", + " one_hot_Y[xp.arange(Y.size), Y] = 1\n", + " return one_hot_Y.T\n", "\n", "def backward_prop(AL, Y, caches):\n", " grads = {}\n", @@ -120,17 +175,17 @@ "\n", " dAL = AL - Y\n", " current_cache = caches[L-1]\n", - " grads[f\"dW{L}\"] = 1 / m * np.dot(dAL, current_cache[0].T)\n", - " grads[f\"db{L}\"] = 1 / m * np.sum(dAL, axis=1, keepdims=True)\n", - " dA_prev = np.dot(current_cache[1].T, dAL)\n", + " grads[f\"dW{L}\"] = 1 / m * xp.dot(dAL, current_cache[0].T)\n", + " grads[f\"db{L}\"] = 1 / m * xp.sum(dAL, axis=1, keepdims=True)\n", + " dA_prev = xp.dot(current_cache[1].T, dAL)\n", "\n", " for l in reversed(range(L-1)):\n", " current_cache = caches[l]\n", " dZ = dA_prev * ReLU_deriv(current_cache[3])\n", - " grads[f\"dW{l+1}\"] = 1 / m * np.dot(dZ, current_cache[0].T)\n", - " grads[f\"db{l+1}\"] = 1 / m * np.sum(dZ, axis=1, keepdims=True)\n", + " grads[f\"dW{l+1}\"] = 1 / m * xp.dot(dZ, current_cache[0].T)\n", + " grads[f\"db{l+1}\"] = 1 / m * xp.sum(dZ, axis=1, keepdims=True)\n", " if l > 0:\n", - " dA_prev = np.dot(current_cache[1].T, dZ)\n", + " dA_prev = xp.dot(current_cache[1].T, dZ)\n", "\n", " return grads\n", "\n", @@ -144,15 +199,15 @@ " return params\n", "\n", "def get_predictions(AL):\n", - " return np.argmax(AL, axis=0)\n", + " return xp.argmax(AL, axis=0)\n", "\n", "def get_accuracy(predictions, Y):\n", - " return np.sum(predictions == Y) / Y.size" + " return xp.sum(predictions == Y) / Y.size" ] }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ @@ -191,7 +246,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 9, "metadata": {}, "outputs": [], "source": [ @@ -210,7 +265,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 10, "metadata": {}, "outputs": [], "source": [ @@ -222,7 +277,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ @@ -233,9 +288,36 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Performing grid search...\n", + "Training architecture: [1024, 64, 64, 61]\n" + ] + }, + { + "ename": "TypeError", + "evalue": "'ndarray' object cannot be interpreted as an integer", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[12], line 3\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# Perform grid search\u001b[39;00m\n\u001b[1;32m 2\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPerforming grid search...\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m----> 3\u001b[0m best_configs \u001b[38;5;241m=\u001b[39m \u001b[43mgrid_search\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mX_val\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY_val\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlayer_configs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43malpha\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m0.01\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43miterations\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m4000\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m 5\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124mTop 5 Architectures:\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 6\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m config, accuracy, _, _ \u001b[38;5;129;01min\u001b[39;00m best_configs[:\u001b[38;5;241m5\u001b[39m]:\n", + "Cell \u001b[0;32mIn[9], line 7\u001b[0m, in \u001b[0;36mgrid_search\u001b[0;34m(X_train, Y_train, X_val, Y_val, layer_configs, alpha, iterations, accuracy_threshold)\u001b[0m\n\u001b[1;32m 5\u001b[0m layer_dims \u001b[38;5;241m=\u001b[39m [input_size] \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mlist\u001b[39m(layer_config) \u001b[38;5;241m+\u001b[39m [output_size]\n\u001b[1;32m 6\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mTraining architecture: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mlayer_dims\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m----> 7\u001b[0m best_params, accuracy, acc_store \u001b[38;5;241m=\u001b[39m \u001b[43mgradient_descent\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mX_val\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY_val\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlayer_dims\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43malpha\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43miterations\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maccuracy_threshold\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 8\u001b[0m results\u001b[38;5;241m.\u001b[39mappend((layer_config, accuracy, best_params, acc_store))\n\u001b[1;32m 9\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mArchitecture \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mlayer_dims\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m: Best Validation Accuracy: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00maccuracy\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m.4f\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n", + "Cell \u001b[0;32mIn[8], line 8\u001b[0m, in \u001b[0;36mgradient_descent\u001b[0;34m(X_train, Y_train, X_val, Y_val, layer_dims, alpha, iterations, accuracy_threshold)\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(iterations):\n\u001b[1;32m 7\u001b[0m AL, caches \u001b[38;5;241m=\u001b[39m forward_prop(X_train, params)\n\u001b[0;32m----> 8\u001b[0m grads \u001b[38;5;241m=\u001b[39m \u001b[43mbackward_prop\u001b[49m\u001b[43m(\u001b[49m\u001b[43mAL\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcaches\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 9\u001b[0m params \u001b[38;5;241m=\u001b[39m update_params(params, grads, alpha)\n\u001b[1;32m 11\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m i \u001b[38;5;241m%\u001b[39m \u001b[38;5;241m100\u001b[39m \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m0\u001b[39m:\n", + "Cell \u001b[0;32mIn[7], line 42\u001b[0m, in \u001b[0;36mbackward_prop\u001b[0;34m(AL, Y, caches)\u001b[0m\n\u001b[1;32m 40\u001b[0m L \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlen\u001b[39m(caches)\n\u001b[1;32m 41\u001b[0m m \u001b[38;5;241m=\u001b[39m AL\u001b[38;5;241m.\u001b[39mshape[\u001b[38;5;241m1\u001b[39m]\n\u001b[0;32m---> 42\u001b[0m Y \u001b[38;5;241m=\u001b[39m \u001b[43mone_hot\u001b[49m\u001b[43m(\u001b[49m\u001b[43mY\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 44\u001b[0m dAL \u001b[38;5;241m=\u001b[39m AL \u001b[38;5;241m-\u001b[39m Y\n\u001b[1;32m 45\u001b[0m current_cache \u001b[38;5;241m=\u001b[39m caches[L\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m]\n", + "Cell \u001b[0;32mIn[7], line 33\u001b[0m, in \u001b[0;36mone_hot\u001b[0;34m(Y)\u001b[0m\n\u001b[1;32m 32\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mone_hot\u001b[39m(Y):\n\u001b[0;32m---> 33\u001b[0m one_hot_Y \u001b[38;5;241m=\u001b[39m \u001b[43mxp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mzeros\u001b[49m\u001b[43m(\u001b[49m\u001b[43m(\u001b[49m\u001b[43mY\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msize\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mY\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmax\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 34\u001b[0m one_hot_Y[xp\u001b[38;5;241m.\u001b[39marange(Y\u001b[38;5;241m.\u001b[39msize), Y] \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[1;32m 35\u001b[0m one_hot_Y \u001b[38;5;241m=\u001b[39m one_hot_Y\u001b[38;5;241m.\u001b[39mT\n", + "File \u001b[0;32m~/.pyenv/versions/semantics/lib/python3.12/site-packages/cupy/_creation/basic.py:248\u001b[0m, in \u001b[0;36mzeros\u001b[0;34m(shape, dtype, order)\u001b[0m\n\u001b[1;32m 229\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mzeros\u001b[39m(\n\u001b[1;32m 230\u001b[0m shape: _ShapeLike,\n\u001b[1;32m 231\u001b[0m dtype: DTypeLike \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mfloat\u001b[39m,\n\u001b[1;32m 232\u001b[0m order: _OrderCF \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mC\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m 233\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m NDArray[Any]:\n\u001b[1;32m 234\u001b[0m \u001b[38;5;250m \u001b[39m\u001b[38;5;124;03m\"\"\"Returns a new array of given shape and dtype, filled with zeros.\u001b[39;00m\n\u001b[1;32m 235\u001b[0m \n\u001b[1;32m 236\u001b[0m \u001b[38;5;124;03m Args:\u001b[39;00m\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 246\u001b[0m \n\u001b[1;32m 247\u001b[0m \u001b[38;5;124;03m \"\"\"\u001b[39;00m\n\u001b[0;32m--> 248\u001b[0m a \u001b[38;5;241m=\u001b[39m \u001b[43mcupy\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mndarray\u001b[49m\u001b[43m(\u001b[49m\u001b[43mshape\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdtype\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43morder\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43morder\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 249\u001b[0m a\u001b[38;5;241m.\u001b[39mdata\u001b[38;5;241m.\u001b[39mmemset_async(\u001b[38;5;241m0\u001b[39m, a\u001b[38;5;241m.\u001b[39mnbytes)\n\u001b[1;32m 250\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m a\n", + "File \u001b[0;32mcupy/_core/core.pyx:137\u001b[0m, in \u001b[0;36mcupy._core.core.ndarray.__new__\u001b[0;34m()\u001b[0m\n", + "File \u001b[0;32mcupy/_core/core.pyx:202\u001b[0m, in \u001b[0;36mcupy._core.core._ndarray_base._init\u001b[0;34m()\u001b[0m\n", + "\u001b[0;31mTypeError\u001b[0m: 'ndarray' object cannot be interpreted as an integer" + ] + } + ], "source": [ "# Perform grid search\n", "print(\"Performing grid search...\")\n",