{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Boxplot der Accuracy[in %] jedes normalen Datensatzes aus CSV-File" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "fillcolor": "rgba(93, 164, 214, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "CP", "type": "box", "whiskerwidth": 0.2, "y": [ 94.12, 100, 100, 100, 94.12, 100, 97.06, 100, 100 ] }, { "fillcolor": "rgba(255, 144, 14, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AMI", "type": "box", "whiskerwidth": 0.2, "y": [ 78, 74.5, 89, 83.5, 79.5, 85.5, 75, 74, 71.8 ] }, { "fillcolor": "rgba(44, 160, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AWE", "type": "box", "whiskerwidth": 0.2, "y": [ 24.13, 23, 46, 37, 25.13, 23, 38.5, 23.15, 47.5 ] }, { "fillcolor": "rgba(255, 65, 54, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "EarVN_1_0", "type": "box", "whiskerwidth": 0.2, "y": [ 68.14, 79.31, 84.31, 82.27, 38.75, 72.05, 81.31, 70.56, 89.15 ] }, { "fillcolor": "rgba(207, 114, 255, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "UERC", "type": "box", "whiskerwidth": 0.2, "y": [ 43.93, 13.87, 20.26, 13.69, 21.46, 28.46, 34.97, 25.14, 29.56 ] } ], "layout": { "autosize": true, "margin": { "b": 80, "l": 40, "r": 30, "t": 100 }, "paper_bgcolor": "rgb(243, 243, 243)", "plot_bgcolor": "rgb(243, 243, 243)", "showlegend": false, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "title": { "text": "Boxplot der normalen Datensätze mit unterschiedlichen Netzwerken" }, "xaxis": { "autorange": true, "range": [ -0.5, 4.5 ], "title": { "text": "Datensätze (normal)" }, "type": "category" }, "yaxis": { "autorange": true, "dtick": 5, "gridcolor": "rgb(255, 255, 255)", "gridwidth": 1, "range": [ 8.895, 104.795 ], "showgrid": true, "title": { "text": "Accuracy in %" }, "type": "linear", "zeroline": true, "zerolinecolor": "rgb(255, 255, 255)", "zerolinewidth": 2 } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Import Liberies\n", "import plotly.express as px\n", "import plotly.graph_objects as go\n", "import numpy as np\n", "import pandas as pd\n", "\n", "## Read CSV-File with Pandas\n", "data_frame = pd.read_csv(\"CSV_Files/Logbuch_CSV_corr.csv\", sep=';')\n", "\n", "## Array for all collected Accuracies\n", "percentage_all = []\n", "\n", "## Array of all used Datasets\n", "datadir_boxplot = ['CP', 'AMI', 'AWE', 'EarVN_1_0', 'UERC']\n", "\n", "## Read every Accuracy for each Dataset\n", "for j in datadir_boxplot:\n", " Frame = data_frame[data_frame.Datensatz == j]\n", " percentage = []\n", " for i in Frame.Accuracy:\n", " Frame = data_frame[data_frame.Datensatz == str(datadir_boxplot)]\n", " percentage.append(float(i))\n", " percentage_all.append(percentage)\n", " \n", " ## Calculate Mean of each Dataset\n", " #percentage_mean = np.sum(percentage)/9\n", " #print(f':{percentage_mean:0.2f}')\n", "\n", "\n", "y_data = percentage_all\n", "\n", "## Define color for each Dataset\n", "colors = ['rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)', 'rgba(44, 160, 101, 0.5)',\n", " 'rgba(255, 65, 54, 0.5)', 'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)']\n", "\n", "fig = go.Figure()\n", "\n", "\n", "## Define settings of Boxplot-Diagrams\n", "for xd, yd, cls in zip(datadir_boxplot, y_data, colors):\n", " fig.add_trace(go.Box(\n", " y=yd,\n", " name=xd,\n", " jitter=0.5,\n", " whiskerwidth=0.2,\n", " fillcolor=cls,\n", " marker_size=2,\n", " line_width=1)\n", " )\n", "\n", "## Define Layout of Boxplot-Diagrams \n", "fig.update_layout(\n", " title='Boxplot der normalen Datensätze mit unterschiedlichen Netzwerken',\n", " xaxis=dict(\n", " title='Datensätze (normal)',\n", " ),\n", " yaxis=dict(\n", " title='Accuracy in %',\n", " autorange=True,\n", " showgrid=True,\n", " zeroline=True,\n", " dtick=5,\n", " gridcolor='rgb(255, 255, 255)',\n", " gridwidth=1,\n", " zerolinecolor='rgb(255, 255, 255)',\n", " zerolinewidth=2,\n", " ),\n", " margin=dict(\n", " l=40,\n", " r=30,\n", " b=80,\n", " t=100,\n", " ),\n", " paper_bgcolor='rgb(243, 243, 243)',\n", " plot_bgcolor='rgb(243, 243, 243)',\n", " showlegend=False\n", ")\n", "\n", "## Show final Boxplot figure\n", "fig.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Boxplot der Accuracy[in %] jedes erweiterten Datensatzes aus CSV-File" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "fillcolor": "rgba(93, 164, 214, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "CP_e", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 100, 100, 100, 100, 100, 100, 98.53, 100 ] }, { "fillcolor": "rgba(255, 144, 14, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AMI_e", "type": "box", "whiskerwidth": 0.2, "y": [ 98.5, 98.5, 99, 99.75, 98.5, 100, 99.5, 99.5, 100 ] }, { "fillcolor": "rgba(44, 160, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AWE_e", "type": "box", "whiskerwidth": 0.2, "y": [ 42.75, 56.75, 66, 71, 77.63, 93.88, 81.63, 84.88, 98.38 ] }, { "fillcolor": "rgba(255, 65, 54, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "EarVN_1_0", "type": "box", "whiskerwidth": 0.2, "y": [ 68.14, 79.31, 84.31, 82.27, 38.75, 72.05, 81.31, 70.56, 89.15 ] }, { "fillcolor": "rgba(207, 114, 255, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "UERC_e", "type": "box", "whiskerwidth": 0.2, "y": [ 38.55, 46.26, 41.51, 46.17, 75.23, 85.81, 73.54, 85.58, 95.99 ] } ], "layout": { "autosize": true, "margin": { "b": 80, "l": 40, "r": 30, "t": 100 }, "paper_bgcolor": "rgb(243, 243, 243)", "plot_bgcolor": "rgb(243, 243, 243)", "showlegend": false, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "title": { "text": "Boxplot der erweiterten Datensätze mit unterschiedlichen Netzwerken" }, "xaxis": { "autorange": true, "range": [ -0.5, 4.5 ], "title": { "text": "Datensätze (erweitert)" }, "type": "category" }, "yaxis": { "autorange": true, "dtick": 5, "gridcolor": "rgb(255, 255, 255)", "gridwidth": 1, "range": [ 35.136111111111106, 103.41388888888889 ], "showgrid": true, "title": { "text": "Accuracy in %" }, "type": "linear", "zeroline": true, "zerolinecolor": "rgb(255, 255, 255)", "zerolinewidth": 2 } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Import Liberies\n", "import plotly.express as px\n", "import plotly.graph_objects as go\n", "import numpy as np\n", "import pandas as pd\n", "\n", "## Read CSV-File with Pandas\n", "data_frame = pd.read_csv(\"CSV_Files/Logbuch_CSV_corr.csv\", sep=';')\n", "\n", "## Array for all collected Accuracies\n", "percentage_all = []\n", "\n", "## Array of all used Datasets\n", "datadir_boxplot = ['CP_e', 'AMI_e', 'AWE_e', 'EarVN_1_0', 'UERC_e']\n", "\n", "## Read every Accuracy for each network\n", "for j in datadir_boxplot:\n", " Frame = data_frame[data_frame.Datensatz == j]\n", " percentage = []\n", " for i in Frame.Accuracy:\n", " Frame = data_frame[data_frame.Datensatz == str(datadir_boxplot)]\n", " percentage.append(float(i))\n", " percentage_all.append(percentage)\n", " \n", " ## Calculate Mean of every Dataset\n", " #percentage_mean = np.sum(percentage)/9\n", " #print(f':{percentage_mean:0.2f}')\n", "\n", "\n", "y_data = percentage_all\n", "\n", "## Define color for each Dataset\n", "colors = ['rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)', 'rgba(44, 160, 101, 0.5)',\n", " 'rgba(255, 65, 54, 0.5)', 'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)']\n", "\n", "fig = go.Figure()\n", "\n", "\n", "## Define settings of Boxplot-Diagrams\n", "for xd, yd, cls in zip(datadir_boxplot, y_data, colors):\n", " fig.add_trace(go.Box(\n", " y=yd,\n", " name=xd,\n", " jitter=0.5,\n", " whiskerwidth=0.2,\n", " fillcolor=cls,\n", " marker_size=2,\n", " line_width=1)\n", " )\n", "\n", "## Define Layout of Boxplot-Diagrams \n", "fig.update_layout(\n", " title='Boxplot der erweiterten Datensätze mit unterschiedlichen Netzwerken',\n", " xaxis=dict(\n", " title='Datensätze (erweitert)',\n", " ),\n", " yaxis=dict(\n", " title='Accuracy in %',\n", " autorange=True,\n", " showgrid=True,\n", " zeroline=True,\n", " dtick=5,\n", " gridcolor='rgb(255, 255, 255)',\n", " gridwidth=1,\n", " zerolinecolor='rgb(255, 255, 255)',\n", " zerolinewidth=2,\n", " ),\n", " margin=dict(\n", " l=40,\n", " r=30,\n", " b=80,\n", " t=100,\n", " ),\n", " paper_bgcolor='rgb(243, 243, 243)',\n", " plot_bgcolor='rgb(243, 243, 243)',\n", " showlegend=False\n", ")\n", "\n", "## Show final Boxplot figure\n", "fig.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Boxplot der Accuracy[in %] jeder Netzwerke (normale Datensätze)" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "fillcolor": "rgba(93, 164, 214, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "VGG11", "type": "box", "whiskerwidth": 0.2, "y": [ 24.13, 94.12, 78, 68.14, 43.93 ] }, { "fillcolor": "rgba(255, 144, 14, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "VGG11bn", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 74.5, 23, 79.31, 13.87 ] }, { "fillcolor": "rgba(44, 160, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "ResNet18", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 89, 46, 20.26, 84.31 ] }, { "fillcolor": "rgba(90, 16, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "ResNet34", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 83.5, 37, 82.27, 13.69 ] }, { "fillcolor": "rgba(160, 44, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AlexNet", "type": "box", "whiskerwidth": 0.2, "y": [ 94.12, 79.5, 25.13, 38.75, 21.46 ] }, { "fillcolor": "rgba(200, 160, 10, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "SqueezeNet-1-0", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 85.5, 23, 72.05, 28.46 ] }, { "fillcolor": "rgba(255, 65, 54, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "GoogLeNet", "type": "box", "whiskerwidth": 0.2, "y": [ 97.06, 75, 38.5, 81.31, 34.97 ] }, { "fillcolor": "rgba(207, 114, 255, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "Shufflenet-v2-x0_5", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 74, 23.15, 70.56, 25.14 ] }, { "fillcolor": "rgba(127, 96, 0, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "Resnext101-32x8d", "type": "box", "whiskerwidth": 0.2, "y": [ 100, 71.8, 47.5, 89.15, 29.56 ] } ], "layout": { "autosize": true, "margin": { "b": 80, "l": 40, "r": 30, "t": 100 }, "paper_bgcolor": "rgb(243, 243, 243)", "plot_bgcolor": "rgb(243, 243, 243)", "showlegend": false, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "title": { "text": "Boxplot der Netzwerke mit normalen Datensätzen" }, "xaxis": { "autorange": false, "range": [ -0.5, 8.5 ], "title": { "text": "Netzwerke" }, "type": "category" }, "yaxis": { "autorange": true, "dtick": 5, "gridcolor": "rgb(255, 255, 255)", "gridwidth": 1, "range": [ 8.895, 104.795 ], "showgrid": true, "title": { "text": "Accuracy in %" }, "type": "linear", "zeroline": true, "zerolinecolor": "rgb(255, 255, 255)", "zerolinewidth": 2 } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Import Liberies\n", "import plotly.express as px\n", "import plotly.graph_objects as go\n", "import numpy as np\n", "import pandas as pd\n", "\n", "## Read CSV-File with Pandas\n", "data_frame = pd.read_csv(\"CSV_Files/Logbuch_CSV_corr.csv\", sep=';')\n", "\n", "## Array for all collected Accuracies\n", "percentage_all_networks = []\n", "\n", "## Array of all used networks\n", "networks_boxplot = ['VGG11', 'VGG11bn', 'ResNet18', 'ResNet34', 'AlexNet', 'SqueezeNet-1-0', 'GoogLeNet', 'Shufflenet-v2-x0_5', 'Resnext101-32x8d']\n", "\n", "## Read every Accuracy for each network\n", "for j in networks_boxplot:\n", " Frame = data_frame[data_frame.Netzwerk == j]\n", " percentage = []\n", " for i in Frame.Accuracy:\n", " Frame = data_frame[data_frame.Netzwerk == str(networks_boxplot)]\n", " percentage.append(float(i))\n", " percentage_all_networks.append(percentage)\n", " \n", " ## Calculate Mean of every Dataset\n", " #percentage_mean = np.sum(percentage)/5\n", " #print(f'{j}:\\t{percentage_mean:0.2f}')\n", "\n", " \n", "y_data = percentage_all_networks\n", "\n", "## Define color of each Network\n", "colors = ['rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)', 'rgba(44, 160, 101, 0.5)', 'rgba(90, 16, 101, 0.5)', 'rgba(160, 44, 101, 0.5)', 'rgba(200, 160, 10, 0.5)',\n", " 'rgba(255, 65, 54, 0.5)', 'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)']\n", "\n", "fig = go.Figure()\n", "\n", "## Define settings of Boxplot-Diagrams\n", "for xd, yd, cls in zip(networks_boxplot, y_data, colors):\n", " fig.add_trace(go.Box(\n", " y=yd,\n", " name=xd,\n", " jitter=0.5,\n", " whiskerwidth=0.2,\n", " fillcolor=cls,\n", " marker_size=2,\n", " line_width=1)\n", " )\n", "\n", "## Define Layout of Boxplot-Diagrams\n", "fig.update_layout(\n", " title='Boxplot der Netzwerke mit normalen Datensätzen',\n", " xaxis=dict(\n", " title='Netzwerke',\n", " ),\n", " yaxis=dict(\n", " title='Accuracy in %',\n", " autorange=True,\n", " showgrid=True,\n", " zeroline=True,\n", " dtick=5,\n", " gridcolor='rgb(255, 255, 255)',\n", " gridwidth=1,\n", " zerolinecolor='rgb(255, 255, 255)',\n", " zerolinewidth=2,\n", " ),\n", " margin=dict(\n", " l=40,\n", " r=30,\n", " b=80,\n", " t=100,\n", " ),\n", " paper_bgcolor='rgb(243, 243, 243)',\n", " plot_bgcolor='rgb(243, 243, 243)',\n", " showlegend=False\n", ")\n", "\n", "\n", "## Show final Boxplot figure\n", "fig.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Boxplot der Accuracy[in %] jeder Netzwerke (erweiterte Datensätze)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "fillcolor": "rgba(93, 164, 214, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "VGG11_e", "type": "box", "whiskerwidth": 0.2, "y": [ 68.14, 98.5, 100, 42.75, 38.55 ] }, { "fillcolor": "rgba(255, 144, 14, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "VGG11bn_e", "type": "box", "whiskerwidth": 0.2, "y": [ 79.31, 98.5, 100, 56.75, 46.26 ] }, { "fillcolor": "rgba(44, 160, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "ResNet18_e", "type": "box", "whiskerwidth": 0.2, "y": [ 84.31, 99, 100, 66, 41.51 ] }, { "fillcolor": "rgba(90, 16, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "ResNet34_e", "type": "box", "whiskerwidth": 0.2, "y": [ 82.27, 99.75, 100, 71, 46.17 ] }, { "fillcolor": "rgba(160, 44, 101, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "AlexNet_e", "type": "box", "whiskerwidth": 0.2, "y": [ 38.75, 98.5, 100, 77.63, 75.23 ] }, { "fillcolor": "rgba(200, 160, 10, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "SqueezeNet-1-0_e", "type": "box", "whiskerwidth": 0.2, "y": [ 72.05, 100, 100, 93.88, 85.81 ] }, { "fillcolor": "rgba(255, 65, 54, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "GoogLeNet_e", "type": "box", "whiskerwidth": 0.2, "y": [ 81.31, 99.5, 100, 81.63, 73.54 ] }, { "fillcolor": "rgba(207, 114, 255, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "Shufflenet-v2-x0_5_e", "type": "box", "whiskerwidth": 0.2, "y": [ 70.56, 99.5, 98.53, 84.88, 85.58 ] }, { "fillcolor": "rgba(127, 96, 0, 0.5)", "jitter": 0.5, "line": { "width": 1 }, "marker": { "size": 2 }, "name": "Resnext101-32x8d_e", "type": "box", "whiskerwidth": 0.2, "y": [ 89.15, 100, 100, 98.38, 95.99 ] } ], "layout": { "autosize": true, "margin": { "b": 80, "l": 40, "r": 30, "t": 100 }, "paper_bgcolor": "rgb(243, 243, 243)", "plot_bgcolor": "rgb(243, 243, 243)", "showlegend": false, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "title": { "text": "Boxplot der Netzwerke mit erweiterten Datensätzen" }, "xaxis": { "autorange": true, "range": [ -0.5, 8.5 ], "title": { "text": "Netzwerke" }, "type": "category" }, "yaxis": { "autorange": true, "dtick": 5, "gridcolor": "rgb(255, 255, 255)", "gridwidth": 1, "range": [ 35.136111111111106, 103.41388888888889 ], "showgrid": true, "title": { "text": "Accuracy in %" }, "type": "linear", "zeroline": true, "zerolinecolor": "rgb(255, 255, 255)", "zerolinewidth": 2 } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Import Liberies\n", "import plotly.express as px\n", "import plotly.graph_objects as go\n", "import numpy as np\n", "import pandas as pd\n", "\n", "## Read CSV-File with Pandas\n", "data_frame = pd.read_csv(\"CSV_Files/Logbuch_CSV_corr.csv\", sep=';')\n", "\n", "## Array for all collected Accuracies\n", "percentage_all_networks = []\n", "\n", "## Array of all used networks\n", "networks_boxplot = ['VGG11_e', 'VGG11bn_e', 'ResNet18_e', 'ResNet34_e', 'AlexNet_e', 'SqueezeNet-1-0_e', 'GoogLeNet_e', 'Shufflenet-v2-x0_5_e', 'Resnext101-32x8d_e']\n", "\n", "## Read every Accuracy of each Network\n", "for j in networks_boxplot:\n", " Frame = data_frame[data_frame.Netzwerk == j]\n", " percentage = []\n", " for i in Frame.Accuracy:\n", " Frame = data_frame[data_frame.Netzwerk == str(networks_boxplot)]\n", " percentage.append(float(i))\n", " percentage_all_networks.append(percentage)\n", " \n", " ## Calculate Mean of each Network\n", " #percentage_mean = np.sum(percentage)/5\n", " ##print(f'{j}:\\t\\t{percentage_mean:0.2f}')\n", "\n", "\n", "y_data = percentage_all_networks\n", "\n", "## Define color of each Network\n", "colors = ['rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)', 'rgba(44, 160, 101, 0.5)', 'rgba(90, 16, 101, 0.5)', 'rgba(160, 44, 101, 0.5)', 'rgba(200, 160, 10, 0.5)',\n", " 'rgba(255, 65, 54, 0.5)', 'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)']\n", "\n", "fig = go.Figure()\n", "\n", "## Define settings of Boxplot-Diagrams\n", "for xd, yd, cls in zip(networks_boxplot, y_data, colors):\n", " fig.add_trace(go.Box(\n", " y=yd,\n", " name=xd,\n", " jitter=0.5,\n", " whiskerwidth=0.2,\n", " fillcolor=cls,\n", " marker_size=2,\n", " line_width=1)\n", " )\n", "\n", "## Define Layout of Boxplot-Diagrams\n", "fig.update_layout(\n", " title='Boxplot der Netzwerke mit erweiterten Datensätzen',\n", " xaxis=dict(\n", " title='Netzwerke',\n", " ),\n", " yaxis=dict(\n", " title='Accuracy in %',\n", " autorange=True,\n", " showgrid=True,\n", " zeroline=True,\n", " dtick=5,\n", " gridcolor='rgb(255, 255, 255)',\n", " gridwidth=1,\n", " zerolinecolor='rgb(255, 255, 255)',\n", " zerolinewidth=2,\n", " ),\n", " margin=dict(\n", " l=40,\n", " r=30,\n", " b=80,\n", " t=100,\n", " ),\n", " paper_bgcolor='rgb(243, 243, 243)',\n", " plot_bgcolor='rgb(243, 243, 243)',\n", " showlegend=False\n", ")\n", "\n", "\n", "## Show final Boxplot figure\n", "#fig.show()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.8.3" } }, "nbformat": 4, "nbformat_minor": 4 }