{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Generated Recipes for evaluation"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import sys\n",
"sys.path.append(\"../\")\n",
"sys.path.append(\"../EvolutionaryAlgorithm/\")"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"../EvolutionaryAlgorithm/EvolutionaryAlgorithm.py:60: TqdmExperimentalWarning:\n",
"\n",
"Using `tqdm.autonotebook.tqdm` in notebook mode. Use `tqdm.tqdm` instead to force console mode (e.g. in jupyter console)\n",
"\n"
]
}
],
"source": [
"import EvolutionaryAlgorithm as EA"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ff3e966495934c2aa709552707de2b4d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9ff44ec2409b4fc28a890e59a49e4a88",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ec252693c93c447e965f643327308a1d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5196b4ed2472483189f37f8fe609ee7d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "35085bd93fcb4f56b3e0e1580c2ac3ac",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c1cf78ab6a44435e85256371868aadc2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "35e5e4e9930046729edb60999a2eb217",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "17fa2f6a52264986aa0b5a93ecc3a2c3",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d8201b09b3054dfe9bbb8483900a5d6e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "75f8cff69b7548338289ef9b97d955d1",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot mutate given node\n",
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot mutate given node\n",
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot find matching action node for mutation\n",
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5a8635c7c7cf42acb0ba82ae3c3ee93a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "54205229ba31456e8b8c68e3db96d598",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "762a1932d1fa42ebbdd018bbe6f3cce8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "548c081efdad4ede9311919e3b89e2b5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Warning: cannot find matching action node for mutation\n",
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b5f82595ac9f4dcabb946d55a513bb46",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ba073e965d7e4d04a5511202a7a6cd71",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Warning: cannot mutate given node\n",
"Warning: cannot find matching action node for mutation\n",
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "587e80beb2154d04894d9375952bd9e0",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bc0002c9299f48a290fb2fc68916b70f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "41a39aee28c94cf397e88d190209dde8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='build initial population', max=75.0, style=ProgressStyle(…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "feababa007f04609996469c366a24e3b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='run evolutionary cycles', max=35.0, style=ProgressStyle(d…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"ssc = []\n",
"\n",
"for i in range(10):\n",
"\n",
" p = EA.Population([\"rice\",],['rice'], min_additional=5, max_additional=11, n_population = 75, mutations=2)\n",
"\n",
" ssc.append(p.run(35))\n",
"\n",
" scores = [tree.score() for tree in p.population]\n",
"\n",
" ii = np.argsort(-np.array(scores))\n",
"\n",
" samples_i = EA.np.random.choice(ii[:1],1)\n",
"\n",
" samples = [p.population[i] for i in samples_i]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"scores = np.array(ssc)"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [],
"source": [
"aavg = np.average(scores, axis=0)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"