commit 8e0720eab3ee7bfee86271f35e518ba99eed6074
parent 93b876d0e273ad88f339091d27d5caf7ab5c17e8
Author: Andrew <andrewlaack1@gmail.com>
Date: Thu, 7 Nov 2024 19:26:18 -0600
NN sorting seems... possibly a valid avenue of research.
Diffstat:
2 files changed, 391 insertions(+), 1 deletion(-)
diff --git a/decisionTreeClassifierFromScratch/DecisionTreeClassifier.ipynb b/decisionTreeClassifierFromScratch/DecisionTreeClassifier.ipynb
@@ -7073,7 +7073,7 @@
},
{
"cell_type": "code",
- "execution_count": 818,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -7108,6 +7108,46 @@
},
{
"cell_type": "code",
+ "execution_count": 821,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array(['5', '4', '9', '1', '3', '4', '1', '7', '2', '7'], dtype='<U32')"
+ ]
+ },
+ "execution_count": 821,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "customPreds[:10]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 822,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array(['8', '7', '4', '0', '0', '2', '9', '1', '6', '4'], dtype=object)"
+ ]
+ },
+ "execution_count": 822,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "test[:,-1][:10]"
+ ]
+ },
+ {
+ "cell_type": "code",
"execution_count": 820,
"metadata": {},
"outputs": [
diff --git a/deepSorting/NeuralNetworkForSorting.ipynb b/deepSorting/NeuralNetworkForSorting.ipynb
@@ -0,0 +1,350 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 86,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import keras\n",
+ "import numpy as np\n",
+ "\n",
+ "arrs = np.random.rand(1000000, 10)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 87,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "X = arrs.copy()\n",
+ "y = arrs\n",
+ "for i in range(0, len(arrs)):\n",
+ " arrs[i].sort()\n",
+ " y[i] = arrs[i]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 88,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "model = keras.Sequential(layers=[\n",
+ " keras.layers.Input((10,)),\n",
+ " keras.layers.Dense(256, 'relu'),\n",
+ " keras.layers.Dense(256, 'relu'),\n",
+ " keras.layers.Dense(256, 'relu'),\n",
+ " keras.layers.Dense(256, 'relu'),\n",
+ " keras.layers.Dense(256, 'relu'),\n",
+ " keras.layers.Dense(10, 'relu')\n",
+ "])\n",
+ "\n",
+ "model.compile(optimizer='adam', loss='mse')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 89,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Epoch 1/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 0.0856\n",
+ "Epoch 2/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m24s\u001b[0m 2ms/step - loss: 0.0427\n",
+ "Epoch 3/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 0.0426\n",
+ "Epoch 4/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 0.0055\n",
+ "Epoch 5/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 9.9785e-05\n",
+ "Epoch 6/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 8.7118e-05\n",
+ "Epoch 7/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 7.8787e-05\n",
+ "Epoch 8/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 7.3314e-05\n",
+ "Epoch 9/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 6.8400e-05\n",
+ "Epoch 10/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 6.3560e-05\n",
+ "Epoch 11/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 2ms/step - loss: 6.0688e-05\n",
+ "Epoch 12/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 5.7792e-05\n",
+ "Epoch 13/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 5.5196e-05\n",
+ "Epoch 14/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 5.3011e-05\n",
+ "Epoch 15/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 2ms/step - loss: 5.0850e-05\n",
+ "Epoch 16/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.9164e-05\n",
+ "Epoch 17/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.7786e-05\n",
+ "Epoch 18/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.6064e-05\n",
+ "Epoch 19/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.4914e-05\n",
+ "Epoch 20/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.3817e-05\n",
+ "Epoch 21/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.2979e-05\n",
+ "Epoch 22/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.2093e-05\n",
+ "Epoch 23/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.1059e-05\n",
+ "Epoch 24/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 4.0475e-05\n",
+ "Epoch 25/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.9875e-05\n",
+ "Epoch 26/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.9346e-05\n",
+ "Epoch 27/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.8780e-05\n",
+ "Epoch 28/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.8075e-05\n",
+ "Epoch 29/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.7682e-05\n",
+ "Epoch 30/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.7309e-05\n",
+ "Epoch 31/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.6973e-05\n",
+ "Epoch 32/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.6500e-05\n",
+ "Epoch 33/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.6221e-05\n",
+ "Epoch 34/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.5780e-05\n",
+ "Epoch 35/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.5255e-05\n",
+ "Epoch 36/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.4851e-05\n",
+ "Epoch 37/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.4683e-05\n",
+ "Epoch 38/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.4109e-05\n",
+ "Epoch 39/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.3758e-05\n",
+ "Epoch 40/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.3491e-05\n",
+ "Epoch 41/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.3178e-05\n",
+ "Epoch 42/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.2711e-05\n",
+ "Epoch 43/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.2707e-05\n",
+ "Epoch 44/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.2125e-05\n",
+ "Epoch 45/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.2062e-05\n",
+ "Epoch 46/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.1909e-05\n",
+ "Epoch 47/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.1489e-05\n",
+ "Epoch 48/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.1198e-05\n",
+ "Epoch 49/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.1206e-05\n",
+ "Epoch 50/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0994e-05\n",
+ "Epoch 51/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0871e-05\n",
+ "Epoch 52/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0583e-05\n",
+ "Epoch 53/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0572e-05\n",
+ "Epoch 54/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0405e-05\n",
+ "Epoch 55/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 3.0263e-05\n",
+ "Epoch 56/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9958e-05\n",
+ "Epoch 57/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9656e-05\n",
+ "Epoch 58/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9605e-05\n",
+ "Epoch 59/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9368e-05\n",
+ "Epoch 60/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9298e-05\n",
+ "Epoch 61/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9185e-05\n",
+ "Epoch 62/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.9058e-05\n",
+ "Epoch 63/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8837e-05\n",
+ "Epoch 64/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8828e-05\n",
+ "Epoch 65/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8554e-05\n",
+ "Epoch 66/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8403e-05\n",
+ "Epoch 67/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8261e-05\n",
+ "Epoch 68/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7985e-05\n",
+ "Epoch 69/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.8051e-05\n",
+ "Epoch 70/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7925e-05\n",
+ "Epoch 71/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7633e-05\n",
+ "Epoch 72/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7610e-05\n",
+ "Epoch 73/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7255e-05\n",
+ "Epoch 74/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.7215e-05\n",
+ "Epoch 75/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6946e-05\n",
+ "Epoch 76/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6809e-05\n",
+ "Epoch 77/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6713e-05\n",
+ "Epoch 78/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6600e-05\n",
+ "Epoch 79/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6570e-05\n",
+ "Epoch 80/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6319e-05\n",
+ "Epoch 81/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6275e-05\n",
+ "Epoch 82/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5981e-05\n",
+ "Epoch 83/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.6002e-05\n",
+ "Epoch 84/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5906e-05\n",
+ "Epoch 85/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5878e-05\n",
+ "Epoch 86/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5709e-05\n",
+ "Epoch 87/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 2ms/step - loss: 2.5713e-05\n",
+ "Epoch 88/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5431e-05\n",
+ "Epoch 89/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5485e-05\n",
+ "Epoch 90/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5247e-05\n",
+ "Epoch 91/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5124e-05\n",
+ "Epoch 92/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.5110e-05\n",
+ "Epoch 93/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4945e-05\n",
+ "Epoch 94/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4931e-05\n",
+ "Epoch 95/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 2ms/step - loss: 2.4784e-05\n",
+ "Epoch 96/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4788e-05\n",
+ "Epoch 97/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4458e-05\n",
+ "Epoch 98/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4624e-05\n",
+ "Epoch 99/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4258e-05\n",
+ "Epoch 100/100\n",
+ "\u001b[1m10000/10000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 2ms/step - loss: 2.4311e-05\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "<keras.src.callbacks.history.History at 0x7fde9d64eb10>"
+ ]
+ },
+ "execution_count": 89,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "model.fit(batch_size=100, x=X, y=y, epochs=100)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 91,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import keras\n",
+ "import numpy as np\n",
+ "\n",
+ "X_val = np.random.rand(1000, 10)\n",
+ "y_val = X_val.copy()\n",
+ "\n",
+ "for i in range(0, len(y_val)):\n",
+ " y_val[i].sort()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 95,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n"
+ ]
+ }
+ ],
+ "source": [
+ "y_pred = model.predict(X_val)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 100,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "0.0035610408570671526"
+ ]
+ },
+ "execution_count": 100,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from sklearn.metrics import mean_absolute_error\n",
+ "\n",
+ "mean_absolute_error(y_pred=y_pred, y_true=y_val)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": ".venv",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.11.2"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
