python_for_analytics
/
HW1_neural_nets_Shalimov_Roman.ipynb
880 строк · 135.7 Кб
1{2"cells": [3{4"cell_type": "markdown",5"metadata": {6"id": "ra21eO61q-iI"7},8"source": [9"## Задание\n",10"\n",11"Подобрать такие значения параметров модели, чтобы модель давала верный прогноз по всем случаям."12]13},14{15"cell_type": "code",16"execution_count": 2,17"metadata": {18"id": "Ptsy6rKbQU95"19},20"outputs": [],21"source": [22"import pandas as pd\n",23"import numpy as np\n",24"import seaborn as sns\n",25"import matplotlib.pyplot as plt\n",26"\n",27"from ipywidgets import interact, interactive, fixed, interact_manual\n",28"import ipywidgets as widgets\n",29"from IPython.display import display"30]31},32{33"cell_type": "code",34"execution_count": 3,35"metadata": {36"colab": {37"base_uri": "https://localhost:8080/",38"height": 17539},40"id": "TZMlhvY_Qkwf",41"outputId": "891cf4a2-c84f-4c37-ecdd-c8490612f096"42},43"outputs": [44{45"data": {46"text/html": [47"<div>\n",48"<style scoped>\n",49" .dataframe tbody tr th:only-of-type {\n",50" vertical-align: middle;\n",51" }\n",52"\n",53" .dataframe tbody tr th {\n",54" vertical-align: top;\n",55" }\n",56"\n",57" .dataframe thead th {\n",58" text-align: right;\n",59" }\n",60"</style>\n",61"<table border=\"1\" class=\"dataframe\">\n",62" <thead>\n",63" <tr style=\"text-align: right;\">\n",64" <th></th>\n",65" <th>p_1</th>\n",66" <th>p_2</th>\n",67" <th>target</th>\n",68" </tr>\n",69" </thead>\n",70" <tbody>\n",71" <tr>\n",72" <th>0</th>\n",73" <td>1</td>\n",74" <td>1</td>\n",75" <td>0</td>\n",76" </tr>\n",77" <tr>\n",78" <th>1</th>\n",79" <td>1</td>\n",80" <td>0</td>\n",81" <td>1</td>\n",82" </tr>\n",83" <tr>\n",84" <th>2</th>\n",85" <td>0</td>\n",86" <td>1</td>\n",87" <td>1</td>\n",88" </tr>\n",89" <tr>\n",90" <th>3</th>\n",91" <td>0</td>\n",92" <td>0</td>\n",93" <td>0</td>\n",94" </tr>\n",95" </tbody>\n",96"</table>\n",97"</div>"98],99"text/plain": [100" p_1 p_2 target\n",101"0 1 1 0\n",102"1 1 0 1\n",103"2 0 1 1\n",104"3 0 0 0"105]106},107"execution_count": 3,108"metadata": {},109"output_type": "execute_result"110}111],112"source": [113"data = pd.DataFrame([[1,1,0],\n",114" [1,0,1],\n",115" [0,1,1],\n",116" [0,0,0]], columns=['p_1','p_2','target'])\n",117"data.head()"118]119},120{121"cell_type": "markdown",122"metadata": {123"id": "UCpi3verrKVR"124},125"source": [126"## Функция для отрисовки граници разделения классов"127]128},129{130"cell_type": "code",131"execution_count": 3,132"metadata": {133"id": "KzRRCMatgRrr"134},135"outputs": [],136"source": [137"def plot_ex(df, name_x1, name_x2, fun):\n",138" x_min, x_max = df[name_x1].min() - 5, df[name_x1].max() + 5\n",139" y_min, y_max = df[name_x2].min() - 5, df[name_x2].max() + 5\n",140"\n",141" h = .1\n",142" xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))\n",143"\n",144" Z = fun(np.c_[xx.ravel(),yy.ravel()])\n",145" Z = Z.reshape(xx.shape)\n",146" plt.figure(1, figsize=(7,7))\n",147" plt.pcolormesh(xx,yy,Z,cmap=plt.cm.Pastel1)\n",148"\n",149" sns.scatterplot(x=df[name_x1], y=df[name_x2], hue=df['target'], palette=['#00aa00','#0000ff'])\n",150" plt.xlabel(name_x1)\n",151" plt.ylabel(name_x2)\n",152"\n",153" plt.xlim(xx.min(), xx.max())\n",154" plt.ylim(yy.min(), yy.max())\n",155" plt.xticks()\n",156" plt.show()"157]158},159{160"cell_type": "markdown",161"metadata": {162"id": "mR8dHXxEtQo4"163},164"source": [165"## Подберем параметры модели"166]167},168{169"cell_type": "markdown",170"metadata": {171"id": "M5iGSbl1rf7M"172},173"source": [174"\n"175]176},177{178"cell_type": "markdown",179"metadata": {180"id": "L8SmWLQGulVA"181},182"source": [183"### Модель\n",184"Модель состоит извходного слоя с двумя входами, так как у нас всего 2 признака, из промежуточного слоя из трех нейронов и выходного слоя с одним нейроном.\n",185"\n",186"Для промежуточного слоя значения определяются формулами\n",187"\n",188"$f_3 = w_{13} x_1 + w_{23} x_2 + b_3$\n",189"\n",190"$f_4 = w_{14} x_1 + w_{24} x_2 + b_4$\n",191"\n",192"$f_5 = w_{15} x_1 + w_{25} x_2 + b_5$\n",193"\n",194"Функция активация на промежуточных нейронах пороговая\n",195"\n",196"\\begin{align}\n",197" porog(x) = \\left\\{\n",198" \\begin{array}{cl}\n",199" 0 & x \\le 0 \\\\\n",200" 1 & x > 0.\n",201" \\end{array}\n",202" \\right.\n",203"\\end{align}\n",204"Выходой нейрон \n",205"\n",206"$f_6 = w_{36} f_3 + w_{46} f_4 + w_{56} f_5 + b_6$\n",207"\n",208"Функция активация на выходе - сигмоида\n",209"\n",210"$sigmoida(x) = \\frac{1}{1+e^{-x}}$\n",211"\n",212"Порог для принятия решения о том, что объект относится к 1 классу равен 0.5\n",213"\n",214"Итог:\n",215"\n",216"$f_3 = porog(w_{13} x_1 + w_{23} x_2 + b_3)$\n",217"\n",218"$f_4 = porog(w_{14} x_1 + w_{24} x_2 + b_4)$\n",219"\n",220"$f_5 = porog(w_{15} x_1 + w_{25} x_2 + b_5)$\n",221"\n",222"$f(x_1,x_2) = sigmoida(w_{36} f_3 + w_{46} f_4 + w_{56} f_5 + b_6)$\n",223"\n",224"Задача в том, чтобы подобрать такие \n",225"\n",226"$w_{13}$, $w_{14}$, $w_{15}$, $w_{23}$, $w_{24}$, $w_{25}$, $w_{36}$, $w_{46}$, $w_{56}$, $b_3$, $b_4$, $b_5$ \n",227"\n",228"что бы модель давала верный прогноз по всем случаям\n",229"\n",230"По простому нам нужно, что бы точки одного класса (цвета) оказались на одном цветовом поле, а точки другого класса на другом поле.\n"231]232},233{234"cell_type": "code",235"execution_count": 4,236"metadata": {237"colab": {238"base_uri": "https://localhost:8080/",239"height": 845,240"referenced_widgets": [241"07345103480a45a9b0b73693bfa8b865",242"c4df7daba3744362b6861458ec26871a",243"3c36a1420dd844e1a3d465508bc00e6d",244"be96a53872854f0888a509de6372a9c7",245"7e6350aa761247039c9e7fb77857116f",246"6ed5f80e5a6c4ca8978a8a512f95d1ce",247"50b69d91f4f44a018511ab123bf9ca08",248"1764e4e6e122436ba78bb1c919a76fdf",249"30dae3efe92e45b387a7941fcbb6fa27",250"e34ac9e7f31c4b5ebd40b541fc21ab64",251"37ac2b13194d442db3e5527c198d1a46",252"8dbab20ad68a408f93baf2ee95f6b920",253"a35b448372334fa599d8565e595f126f",254"90345fd00ff245c68144f711f1601075",255"07944b1ac3bc4e1b96b29273854e99ca",256"bf41ca5c87b047b698b72b77032088d6",257"57e9de67c11740319f46444fa8d4c6f5",258"ae0acfd3076c4e7db083a09aa6243949",259"80f3db2b23ef414ba0588559dd33bce6",260"36bfe11eb43541dbaa52abc0bc019e90",261"a073bcdd1eb447a7b7057ee7a7d6c6ef",262"4d3201cc20cb4b7786ef0119093aa3be",263"9c4d6069a2c8459b81002242c3de5b55",264"9cbc3560361349afaf8b07a32e75d8dc",265"94e479d1d5ce4eb881dbb402db58b144",266"5d8a8d010c5a40ea9b75453e0cea4f56",267"05a24f1f2914498d8cf579a9393238e2",268"d4a00a60212d42789839aa1104f92c03",269"517deae29cbe4c91afb7d6834e11eaa0",270"3fb86befd4724b5bafa4ba94229dcb7b",271"4863920ad3574922a60e5bdeff833eba",272"30756fe55892483cb41417b003b4cea3",273"66fb9645c6a94a1ebdaa9e2dc6d6d124",274"af516ffdb670438ead818154afc03428",275"909a10e07b514785ad3267cefd338efb",276"bfe124433ac242348ebd837d0bb54fe3",277"4fb15c720a73439db95365cfb1f76129",278"9627e564a7f04e0cbbbca87532398202",279"ab1fa57046c644fc91cb78c26c7f30ac",280"7f1d4e1c77f04556b1ba476e434c8e0f"281]282},283"id": "ABOcMLTwgYFC",284"outputId": "272c284a-171e-4808-a038-945e7be9b194"285},286"outputs": [287{288"data": {289"application/vnd.jupyter.widget-view+json": {290"model_id": "f2a3e5b57617423dab6eec1219232806",291"version_major": 2,292"version_minor": 0293},294"text/plain": [295"interactive(children=(FloatSlider(value=0.0, description='w13', max=10.0, min=-10.0, step=0.05), FloatSlider(v…"296]297},298"metadata": {},299"output_type": "display_data"300}301],302"source": [303"# Подбираем значения параметров модели\n",304"# =====================\n",305"\n",306"# функция описывающая работу нашей модели \n",307"# здесь х это массив пар значений признаков\n",308"# вида [[0.1, 0.2]\n",309"# [1.3, 3.1]\n",310"# [2.1, 1.2]\n",311"# ... ]\n",312"def func(w13, w14, w15, w23, w24, w25, b3, b4, b5, w36, w46,w56):\n",313"\n",314" def sigmoida(x):\n",315" return 1/(1+np.exp(-x))\n",316"\n",317" def porog(x):\n",318" return (x>0)*1\n",319"\n",320" def f(x): \n",321" f3 = porog(w13*x[:,0]+w23*x[:,1]+b3)\n",322" f4 = porog(w14*x[:,0]+w24*x[:,1]+b4)\n",323" f5 = porog(w15*x[:,0]+w25*x[:,1]+b5)\n",324" f6 = w36*f3+w46*f4+w56*f5 \n",325" res = sigmoida(f6)\n",326" return (res<=0.5)*1\n",327"\n",328" # Сделаем прогноз \n",329" data['pred'] = f(data[['p_1','p_2']].to_numpy())\n",330"\n",331" # Нарисуем границу разделения классов и выведем результат предсказания\n",332" plot_ex(data, 'p_1', 'p_2', f)\n",333" print(f\"Доля верных ответов: {sum(data['target'] == data['pred'])/data.shape[0]}\")\n",334"val_range = (-10,10,0.05)\n",335"y=interactive(func, \n",336" w13=val_range, \n",337" w14=val_range, \n",338" w15=val_range, \n",339" w23=val_range, \n",340" w24=val_range, \n",341" w25=val_range, \n",342" b3=val_range, \n",343" b4=val_range, \n",344" b5=val_range,\n",345" w36=val_range, \n",346" w46=val_range, \n",347" w56=val_range, \n",348" )\n",349"display(y)"350]351},352{353"cell_type": "markdown",354"metadata": {},355"source": [356"\n",357"\n",358"Доля верных ответов: 1.0"359]360},361{362"cell_type": "code",363"execution_count": 4,364"metadata": {365"colab": {366"base_uri": "https://localhost:8080/",367"height": 175368},369"id": "eKislGYQ1VN1",370"outputId": "73bdacb6-581f-4a2a-c50e-7e1b1f777928"371},372"outputs": [373{374"data": {375"text/html": [376"<div>\n",377"<style scoped>\n",378" .dataframe tbody tr th:only-of-type {\n",379" vertical-align: middle;\n",380" }\n",381"\n",382" .dataframe tbody tr th {\n",383" vertical-align: top;\n",384" }\n",385"\n",386" .dataframe thead th {\n",387" text-align: right;\n",388" }\n",389"</style>\n",390"<table border=\"1\" class=\"dataframe\">\n",391" <thead>\n",392" <tr style=\"text-align: right;\">\n",393" <th></th>\n",394" <th>p_1</th>\n",395" <th>p_2</th>\n",396" <th>target</th>\n",397" </tr>\n",398" </thead>\n",399" <tbody>\n",400" <tr>\n",401" <th>0</th>\n",402" <td>1</td>\n",403" <td>1</td>\n",404" <td>0</td>\n",405" </tr>\n",406" <tr>\n",407" <th>1</th>\n",408" <td>1</td>\n",409" <td>0</td>\n",410" <td>1</td>\n",411" </tr>\n",412" <tr>\n",413" <th>2</th>\n",414" <td>0</td>\n",415" <td>1</td>\n",416" <td>1</td>\n",417" </tr>\n",418" <tr>\n",419" <th>3</th>\n",420" <td>0</td>\n",421" <td>0</td>\n",422" <td>0</td>\n",423" </tr>\n",424" </tbody>\n",425"</table>\n",426"</div>"427],428"text/plain": [429" p_1 p_2 target\n",430"0 1 1 0\n",431"1 1 0 1\n",432"2 0 1 1\n",433"3 0 0 0"434]435},436"execution_count": 4,437"metadata": {},438"output_type": "execute_result"439}440],441"source": [442"data"443]444},445{446"cell_type": "code",447"execution_count": null,448"metadata": {449"id": "fm2u1yGz2Rko"450},451"outputs": [],452"source": []453}454],455"metadata": {456"colab": {457"provenance": []458},459"kernelspec": {460"display_name": "Python 3 (ipykernel)",461"language": "python",462"name": "python3"463},464"language_info": {465"codemirror_mode": {466"name": "ipython",467"version": 3468},469"file_extension": ".py",470"mimetype": "text/x-python",471"name": "python",472"nbconvert_exporter": "python",473"pygments_lexer": "ipython3",474"version": "3.9.13"475},476"widgets": {477"application/vnd.jupyter.widget-state+json": {478"state": {479"07c6f7bf584e4f4bae598d82416a7506": {480"model_module": "@jupyter-widgets/base",481"model_module_version": "1.2.0",482"model_name": "LayoutModel",483"state": {}484},485"0815bf28053f413288ec46e64de1f9e0": {486"model_module": "@jupyter-widgets/controls",487"model_module_version": "1.5.0",488"model_name": "SliderStyleModel",489"state": {490"description_width": ""491}492},493"0ec90962152341598bcec8ab3c36a66c": {494"model_module": "@jupyter-widgets/controls",495"model_module_version": "1.5.0",496"model_name": "FloatSliderModel",497"state": {498"description": "w25",499"layout": "IPY_MODEL_41726a813b14427a870c0836af5deb66",500"max": 10,501"min": -10,502"step": 0.05,503"style": "IPY_MODEL_f5ad1ddc87764f4aab3dd1fbefc0ec99"504}505},506"12781d5b2b534d8385f0d65d05e54bb6": {507"model_module": "@jupyter-widgets/controls",508"model_module_version": "1.5.0",509"model_name": "FloatSliderModel",510"state": {511"description": "b4",512"layout": "IPY_MODEL_70c3a54e577f43dba1e5d4d2c05f7b10",513"max": 10,514"min": -10,515"step": 0.05,516"style": "IPY_MODEL_511fb8bb835449cca510a4a000b86337",517"value": -0.1518}519},520"17268ddf88704613bb7093726f1c1e9f": {521"model_module": "@jupyter-widgets/controls",522"model_module_version": "1.5.0",523"model_name": "FloatSliderModel",524"state": {525"description": "w46",526"layout": "IPY_MODEL_c9704814117c44c593231890642caba1",527"max": 10,528"min": -10,529"step": 0.05,530"style": "IPY_MODEL_97ec65d3e19b4b1388ff1eb0ff2a1764",531"value": -1532}533},534"1f6101f3790847049dd106dbd5210c66": {535"model_module": "@jupyter-widgets/base",536"model_module_version": "1.2.0",537"model_name": "LayoutModel",538"state": {}539},540"2208abe8ccbe42e299d574c17f5d349e": {541"model_module": "@jupyter-widgets/base",542"model_module_version": "1.2.0",543"model_name": "LayoutModel",544"state": {}545},546"25e2d4c317574b8a8909baa07ebc81df": {547"model_module": "@jupyter-widgets/controls",548"model_module_version": "1.5.0",549"model_name": "SliderStyleModel",550"state": {551"description_width": ""552}553},554"2cb05ef5660f4461a5ec9bdee37e20d9": {555"model_module": "@jupyter-widgets/base",556"model_module_version": "1.2.0",557"model_name": "LayoutModel",558"state": {}559},560"2d27c7a29bde43fdbb6288677a3391bc": {561"model_module": "@jupyter-widgets/base",562"model_module_version": "1.2.0",563"model_name": "LayoutModel",564"state": {}565},566"32dd82e4aa32442686fdf0eba2593fb6": {567"model_module": "@jupyter-widgets/controls",568"model_module_version": "1.5.0",569"model_name": "SliderStyleModel",570"state": {571"description_width": ""572}573},574"33089ae23f1c44e393618d92265c9d6a": {575"model_module": "@jupyter-widgets/base",576"model_module_version": "1.2.0",577"model_name": "LayoutModel",578"state": {}579},580"3396200e5f754398a1e5efefc3eec0b6": {581"model_module": "@jupyter-widgets/controls",582"model_module_version": "1.5.0",583"model_name": "FloatSliderModel",584"state": {585"description": "b3",586"layout": "IPY_MODEL_3d697a7cc11f488f8daeccd4e5dc71f2",587"max": 10,588"min": -10,589"step": 0.05,590"style": "IPY_MODEL_9d588f74101d423ab68a60fb6b4ab58e",591"value": 0.1592}593},594"3d697a7cc11f488f8daeccd4e5dc71f2": {595"model_module": "@jupyter-widgets/base",596"model_module_version": "1.2.0",597"model_name": "LayoutModel",598"state": {}599},600"41726a813b14427a870c0836af5deb66": {601"model_module": "@jupyter-widgets/base",602"model_module_version": "1.2.0",603"model_name": "LayoutModel",604"state": {}605},606"43877c65b5604541941c568a5393f4d1": {607"model_module": "@jupyter-widgets/controls",608"model_module_version": "1.5.0",609"model_name": "FloatSliderModel",610"state": {611"description": "w23",612"layout": "IPY_MODEL_1f6101f3790847049dd106dbd5210c66",613"max": 10,614"min": -10,615"step": 0.05,616"style": "IPY_MODEL_f1bbcc89d6fc4debbe0315ee1b192e02",617"value": -1618}619},620"447301ae809041e98b521d81385bda8b": {621"model_module": "@jupyter-widgets/controls",622"model_module_version": "1.5.0",623"model_name": "FloatSliderModel",624"state": {625"description": "w24",626"layout": "IPY_MODEL_6508cce00bcb44eba96c710b3cc19354",627"max": 10,628"min": -10,629"step": 0.05,630"style": "IPY_MODEL_32dd82e4aa32442686fdf0eba2593fb6",631"value": -1632}633},634"4b136f9cd3fa40a5a878d5c9b6cb9b01": {635"model_module": "@jupyter-widgets/base",636"model_module_version": "1.2.0",637"model_name": "LayoutModel",638"state": {}639},640"511fb8bb835449cca510a4a000b86337": {641"model_module": "@jupyter-widgets/controls",642"model_module_version": "1.5.0",643"model_name": "SliderStyleModel",644"state": {645"description_width": ""646}647},648"59f0802bf700427e995f169421e24f56": {649"model_module": "@jupyter-widgets/base",650"model_module_version": "1.2.0",651"model_name": "LayoutModel",652"state": {}653},654"5ca84006fc874607b7f06b562bccdcc6": {655"model_module": "@jupyter-widgets/output",656"model_module_version": "1.0.0",657"model_name": "OutputModel",658"state": {659"layout": "IPY_MODEL_623cbead830c4e4e9fb7a1f51cff66d3",660"outputs": [661{662"data": {663"image/png": "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\n",664"text/plain": "<Figure size 700x700 with 1 Axes>"665},666"metadata": {},667"output_type": "display_data"668},669{670"name": "stdout",671"output_type": "stream",672"text": "Доля верных ответов: 1.0\n"673}674]675}676},677"623cbead830c4e4e9fb7a1f51cff66d3": {678"model_module": "@jupyter-widgets/base",679"model_module_version": "1.2.0",680"model_name": "LayoutModel",681"state": {}682},683"6508cce00bcb44eba96c710b3cc19354": {684"model_module": "@jupyter-widgets/base",685"model_module_version": "1.2.0",686"model_name": "LayoutModel",687"state": {}688},689"70c3a54e577f43dba1e5d4d2c05f7b10": {690"model_module": "@jupyter-widgets/base",691"model_module_version": "1.2.0",692"model_name": "LayoutModel",693"state": {}694},695"83709371293d48279ea620c5e0ade786": {696"model_module": "@jupyter-widgets/controls",697"model_module_version": "1.5.0",698"model_name": "FloatSliderModel",699"state": {700"description": "w36",701"layout": "IPY_MODEL_33089ae23f1c44e393618d92265c9d6a",702"max": 10,703"min": -10,704"step": 0.05,705"style": "IPY_MODEL_f92dcad19fd94d988a0c986bafa4c4ef",706"value": 1707}708},709"97ec65d3e19b4b1388ff1eb0ff2a1764": {710"model_module": "@jupyter-widgets/controls",711"model_module_version": "1.5.0",712"model_name": "SliderStyleModel",713"state": {714"description_width": ""715}716},717"9af41a07d6ad4bc08a09d8ed8d4dc5f6": {718"model_module": "@jupyter-widgets/controls",719"model_module_version": "1.5.0",720"model_name": "SliderStyleModel",721"state": {722"description_width": ""723}724},725"9d588f74101d423ab68a60fb6b4ab58e": {726"model_module": "@jupyter-widgets/controls",727"model_module_version": "1.5.0",728"model_name": "SliderStyleModel",729"state": {730"description_width": ""731}732},733"a5dbf8b551394c77b2793764ef27ec5e": {734"model_module": "@jupyter-widgets/controls",735"model_module_version": "1.5.0",736"model_name": "SliderStyleModel",737"state": {738"description_width": ""739}740},741"a8bf28b188b241e59d9fd3a702804166": {742"model_module": "@jupyter-widgets/controls",743"model_module_version": "1.5.0",744"model_name": "FloatSliderModel",745"state": {746"description": "w56",747"layout": "IPY_MODEL_4b136f9cd3fa40a5a878d5c9b6cb9b01",748"max": 10,749"min": -10,750"step": 0.05,751"style": "IPY_MODEL_0815bf28053f413288ec46e64de1f9e0"752}753},754"aa4d33d199284d148c4a76eb5c97b0d5": {755"model_module": "@jupyter-widgets/controls",756"model_module_version": "1.5.0",757"model_name": "FloatSliderModel",758"state": {759"description": "w15",760"layout": "IPY_MODEL_2208abe8ccbe42e299d574c17f5d349e",761"max": 10,762"min": -10,763"step": 0.05,764"style": "IPY_MODEL_a5dbf8b551394c77b2793764ef27ec5e"765}766},767"aa520bd859ed4e069462fe4fb16e8ef0": {768"model_module": "@jupyter-widgets/controls",769"model_module_version": "1.5.0",770"model_name": "FloatSliderModel",771"state": {772"description": "w14",773"layout": "IPY_MODEL_2cb05ef5660f4461a5ec9bdee37e20d9",774"max": 10,775"min": -10,776"step": 0.05,777"style": "IPY_MODEL_b7f722d03cca4be692fa440577704e55",778"value": 1779}780},781"afb50357a0504d89a45fa37131c5550a": {782"model_module": "@jupyter-widgets/controls",783"model_module_version": "1.5.0",784"model_name": "FloatSliderModel",785"state": {786"description": "b5",787"layout": "IPY_MODEL_2d27c7a29bde43fdbb6288677a3391bc",788"max": 10,789"min": -10,790"step": 0.05,791"style": "IPY_MODEL_9af41a07d6ad4bc08a09d8ed8d4dc5f6"792}793},794"b7f722d03cca4be692fa440577704e55": {795"model_module": "@jupyter-widgets/controls",796"model_module_version": "1.5.0",797"model_name": "SliderStyleModel",798"state": {799"description_width": ""800}801},802"c9704814117c44c593231890642caba1": {803"model_module": "@jupyter-widgets/base",804"model_module_version": "1.2.0",805"model_name": "LayoutModel",806"state": {}807},808"d6ac6df3b24b41e8b3fcc36651e67324": {809"model_module": "@jupyter-widgets/controls",810"model_module_version": "1.5.0",811"model_name": "FloatSliderModel",812"state": {813"description": "w13",814"layout": "IPY_MODEL_59f0802bf700427e995f169421e24f56",815"max": 10,816"min": -10,817"step": 0.05,818"style": "IPY_MODEL_25e2d4c317574b8a8909baa07ebc81df",819"value": 1.5820}821},822"f1bbcc89d6fc4debbe0315ee1b192e02": {823"model_module": "@jupyter-widgets/controls",824"model_module_version": "1.5.0",825"model_name": "SliderStyleModel",826"state": {827"description_width": ""828}829},830"f2a3e5b57617423dab6eec1219232806": {831"model_module": "@jupyter-widgets/controls",832"model_module_version": "1.5.0",833"model_name": "VBoxModel",834"state": {835"_dom_classes": [836"widget-interact"837],838"children": [839"IPY_MODEL_d6ac6df3b24b41e8b3fcc36651e67324",840"IPY_MODEL_aa520bd859ed4e069462fe4fb16e8ef0",841"IPY_MODEL_aa4d33d199284d148c4a76eb5c97b0d5",842"IPY_MODEL_43877c65b5604541941c568a5393f4d1",843"IPY_MODEL_447301ae809041e98b521d81385bda8b",844"IPY_MODEL_0ec90962152341598bcec8ab3c36a66c",845"IPY_MODEL_3396200e5f754398a1e5efefc3eec0b6",846"IPY_MODEL_12781d5b2b534d8385f0d65d05e54bb6",847"IPY_MODEL_afb50357a0504d89a45fa37131c5550a",848"IPY_MODEL_83709371293d48279ea620c5e0ade786",849"IPY_MODEL_17268ddf88704613bb7093726f1c1e9f",850"IPY_MODEL_a8bf28b188b241e59d9fd3a702804166",851"IPY_MODEL_5ca84006fc874607b7f06b562bccdcc6"852],853"layout": "IPY_MODEL_07c6f7bf584e4f4bae598d82416a7506"854}855},856"f5ad1ddc87764f4aab3dd1fbefc0ec99": {857"model_module": "@jupyter-widgets/controls",858"model_module_version": "1.5.0",859"model_name": "SliderStyleModel",860"state": {861"description_width": ""862}863},864"f92dcad19fd94d988a0c986bafa4c4ef": {865"model_module": "@jupyter-widgets/controls",866"model_module_version": "1.5.0",867"model_name": "SliderStyleModel",868"state": {869"description_width": ""870}871}872},873"version_major": 2,874"version_minor": 0875}876}877},878"nbformat": 4,879"nbformat_minor": 1880}881