Skip to content

Instantly share code, notes, and snippets.

@Saurabh7

Saurabh7/fab-multi.ipynb

Created July 2, 2017 18:41
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save Saurabh7/3b9b96decd8665917a9d0bf89ca73229 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save Saurabh7/3b9b96decd8665917a9d0bf89ca73229 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
fab-multi.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<script>jQuery(function() {if (jQuery(\"body.notebook_app\").length == 0) { jQuery(\".input_area\").toggle(); jQuery(\".prompt\").toggle();}});</script>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"<button onclick=\"jQuery('.input_area').toggle(); jQuery('.prompt').toggle();\">Toggle code</button>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import IPython.core.display as di\n",
"\n",
"# This line will hide code by default when the notebook is exported as HTML\n",
"di.display_html('<script>jQuery(function() {if (jQuery(\"body.notebook_app\").length == 0) { jQuery(\".input_area\").toggle(); jQuery(\".prompt\").toggle();}});</script>', raw=True)\n",
"\n",
"# This line will add a button to toggle visibility of code blocks, for use with the HTML export version\n",
"di.display_html('''<button onclick=\"jQuery('.input_area').toggle(); jQuery('.prompt').toggle();\">Toggle code</button>''', raw=True)"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"from IPython.display import display, HTML\n",
"from itertools import combinations, product\n",
"import pandas as pd\n",
"import json\n",
"from jarvis.brain.utility.data_preparation import get_active_ad_accounts\n",
"from jarvis.brain.insights.paths.path_pre_processor import PathPreProcessor\n",
"from jarvis.brain.insights.paths.path_processor import PathProcessor\n",
"from jarvis.brain.insights.paths.path_utility import create_levels_hash\n",
"from flask.json import jsonify\n",
"import pandas as pd\n",
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"metrics_dict = {\"ga_cpv\": {\"numerator\": \"fb_spend\", \"denominator\":\"ga_visits\", \"diff\": -1},\n",
" \"ga_cpt\": {\"numerator\": \"fb_spend\", \"denominator\":\"ga_transactions\", \"diff\": -1}}\n",
"goal_metric = [\"ga_cpv\", \"ga_cpt\"]\n",
"\n",
"# metrics_dict = {\"fb_ctr\": {\"numerator\": \"fb_inline_link_clicks\", \"denominator\":\"fb_impressions\", \"diff\": 1}}\n",
"# goal_metric = \"fb_ctr\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"args = {\n",
" \"user_id\":\"0\",\n",
" \"account_ids\":{\"facebook\": [261]},\n",
" \"start_date\":\"2017-06-23\",\n",
" \"end_date\":\"2017-06-29\",\n",
" \"goal_metric\": goal_metric,\n",
" \"metrics\": metrics_dict,\n",
" \"num_results\":10,\n",
" \"filters\": [{\"metric\": \"fb_spend\", \"operator\": \"$gt\", \"value\": 0}],\n",
" \"num_trees\": 10,\n",
" \"tag_list\":[\"Sale Campaign\", \"Product Category\",\n",
" \"Image\"],\n",
" \"dimensions\": None,\n",
" \"goal\": None\n",
"}\n",
"\n",
"ad_account = args['account_ids']\n",
"num_levels = 3\n",
"start_date = args['start_date']\n",
"end_date = args['end_date']\n",
"goal_metric = args['goal_metric']\n",
"metrics_dict = args['metrics']\n",
"dimensions = args['dimensions']\n",
"filters = args['filters']\n",
"tag_list = args['tag_list']\n",
"\n",
"aggregations = {}\n",
"for gm in goal_metric:\n",
" aggregations[metrics_dict[gm]['numerator']] = pd.np.sum\n",
" aggregations[metrics_dict[gm]['denominator']] = pd.np.sum\n",
"# aggregations = {metrics_dict[goal_metric]['numerator']: pd.np.sum,\n",
"# metrics_dict[goal_metric]['denominator']: pd.np.sum}\n",
"goal = args['goal']\n",
"metrics = list(aggregations.keys())\n",
"num_trees = args['num_trees'] if args['num_trees'] != None else 10\n",
"\n",
"excluded_tags = ['Campaigns', 'ad_id', 'Adsets', 'start_date', 'Ads', '_id', 'internal_id', 'Sale',\n",
" 'Product - algo', 'Carousel Count', 'Ad Format', 'Dominant Color', 'Number of Faces',\n",
" 'end_date', 'Active Ads Count in Adset', 'Active Adsets Count in Campaign',\n",
" 'Labels', 'audience type', 'audience category', 'Product Set', 'Prominent Text',\n",
" 'DominantColor', 'ProminentText', 'NumberofFaces', 'name', 'strategy', 'Current Adsets',\n",
" 'Facebook Positions', 'AdsetStrategy', 'Publisher Platforms']\n",
"\n",
"num_results = args['num_results'] if args['num_results'] != None else 15\n",
"\n",
"path_pre_process = PathPreProcessor(ad_account, start_date, end_date,\n",
" [], metrics, num_levels, aggregations,\n",
" excluded_tags, filters, dimensions, tag_list=tag_list)\n",
"\n",
"import time\n",
"t1 = time.time()\n",
"path_pre_process.get_data(cross_channel=False)\n",
"print('TIME USED', time.time() - t1)\n",
"path_pre_process.clean_tags(path_pre_process.df, excluded_tags, metrics)\n",
"path_pre_process.score_tags()\n",
"path_pre_process.cluster_tags()\n",
"spend = path_pre_process.get_channel_spend()\n",
"\n",
"global_tag_list = path_pre_process.tag_list\n",
"\n",
"path_model = PathProcessor(path_pre_process.df, goal_metric, metrics_dict, global_tag_list,\n",
" aggregations, num_trees=num_trees, spend=spend)\n",
"# overall_goal_metric=goal)\n",
"path_model.set_comparison(True)\n",
"\n",
"good_df, bad_df, overall = path_model.execute_paths(num_results)\n",
"good_df.replace([pd.np.inf, -pd.np.inf], 0, inplace=True)\n",
"bad_df.replace([pd.np.inf, -pd.np.inf], 0, inplace=True)\n",
"good_df.fillna(value=0, inplace=True)\n",
"bad_df.fillna(value=0, inplace=True)\n",
"\n",
"output = {}\n",
"good_df = good_df[:num_results]\n",
"bad_df = bad_df[:num_results]\n",
"overall_df = good_df.append(bad_df)\n",
"overall_df = overall_df\n",
"\n",
"levels = [('level_' + str(idx)) for idx in range(num_levels)]\n",
"overall_df['levels'] = overall_df[levels].values.tolist()\n",
"overall_df['levels'] = overall_df['levels'].apply(create_levels_hash)\n",
"\n",
"overall_goal_metric = path_model.overall_goal_metrics\n",
"avg_spend_prop = overall_df.spend_prop.mean()\n",
"\n",
"output['goal_axis'] = overall_goal_metric\n",
"output['spend_axis'] = avg_spend_prop\n",
"\n",
"dfs = []\n",
"\n",
"# Separate output for different quadrants.\n",
"dfs.append(overall_df[(overall_df['spend_prop'] < avg_spend_prop) & (\n",
" overall_df[goal_metric] > overall_goal_metric)])\n",
"\n",
"dfs.append(overall_df[(overall_df['spend_prop'] > avg_spend_prop) & (\n",
" overall_df[goal_metric] > overall_goal_metric)])\n",
"\n",
"dfs.append(overall_df[(overall_df['spend_prop'] > avg_spend_prop) & (\n",
" overall_df[goal_metric] < overall_goal_metric)])\n",
"\n",
"dfs.append(overall_df[(overall_df['spend_prop'] < avg_spend_prop) & (\n",
" overall_df[goal_metric] < overall_goal_metric)])\n"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>index</th>\n",
" <th>level_0</th>\n",
" <th>level_1</th>\n",
" <th>level_2</th>\n",
" <th>imp_factor</th>\n",
" <th>ga_cpt</th>\n",
" <th>spend_prop</th>\n",
" <th>ga_cpv</th>\n",
" <th>levels</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0</td>\n",
" <td>Sale Campaign : sale</td>\n",
" <td>Product Category : Indya</td>\n",
" <td>Image : e715ddbac9df21eeae6546df8ce0af83</td>\n",
" <td>4737.765787</td>\n",
" <td>1488.539167</td>\n",
" <td>1.559732</td>\n",
" <td>7.910748</td>\n",
" <td>[{'custom_tags.Sale Campaign': 'sale'}, {'cust...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0</td>\n",
" <td>Image : a1272cba13b15daab9b85c5c3b316fc7</td>\n",
" <td>- : -</td>\n",
" <td>- : -</td>\n",
" <td>83694.715316</td>\n",
" <td>8006.905000</td>\n",
" <td>1.398309</td>\n",
" <td>22.210555</td>\n",
" <td>[{'custom_tags.Image': 'a1272cba13b15daab9b85c...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>3</td>\n",
" <td>Image : f7fe99a0277b16356551860627aa32a6</td>\n",
" <td>- : -</td>\n",
" <td>- : -</td>\n",
" <td>65482.154755</td>\n",
" <td>6835.254000</td>\n",
" <td>2.984236</td>\n",
" <td>16.703944</td>\n",
" <td>[{'custom_tags.Image': 'f7fe99a0277b1635655186...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>5</td>\n",
" <td>Sale Campaign : sale</td>\n",
" <td>Product Category : Multi Category</td>\n",
" <td>- : -</td>\n",
" <td>16641.609226</td>\n",
" <td>3513.827500</td>\n",
" <td>4.909180</td>\n",
" <td>16.807546</td>\n",
" <td>[{'custom_tags.Sale Campaign': 'sale'}, {'cust...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" index level_0 \\\n",
"0 0 Sale Campaign : sale \n",
"0 0 Image : a1272cba13b15daab9b85c5c3b316fc7 \n",
"1 3 Image : f7fe99a0277b16356551860627aa32a6 \n",
"2 5 Sale Campaign : sale \n",
"\n",
" level_1 \\\n",
"0 Product Category : Indya \n",
"0 - : - \n",
"1 - : - \n",
"2 Product Category : Multi Category \n",
"\n",
" level_2 imp_factor ga_cpt \\\n",
"0 Image : e715ddbac9df21eeae6546df8ce0af83 4737.765787 1488.539167 \n",
"0 - : - 83694.715316 8006.905000 \n",
"1 - : - 65482.154755 6835.254000 \n",
"2 - : - 16641.609226 3513.827500 \n",
"\n",
" spend_prop ga_cpv levels \n",
"0 1.559732 7.910748 [{'custom_tags.Sale Campaign': 'sale'}, {'cust... \n",
"0 1.398309 22.210555 [{'custom_tags.Image': 'a1272cba13b15daab9b85c... \n",
"1 2.984236 16.703944 [{'custom_tags.Image': 'f7fe99a0277b1635655186... \n",
"2 4.909180 16.807546 [{'custom_tags.Sale Campaign': 'sale'}, {'cust... "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"overall_df"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"{'ga_cpt': 1657.3468162083975, 'ga_cpv': 10.138876445278633}"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"overall_goal_metric"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"df_1 = dfs[0]\n",
"df_2 = dfs[1]\n",
"df_3 = dfs[2]\n",
"df_4 = dfs[3]\n",
"\n",
"# Change quadrants based on property of goal metric.\n",
"# (e.g. quadrants are exchanged for roas <-> cpt)\n",
"if metrics_dict[goal_metric]['diff'] == -1:\n",
" df_2, df_3 = df_3, df_2\n",
" df_4, df_1 = df_1, df_4"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"from bokeh.charts import Scatter, output_file, show\n",
"from bokeh.io import output_notebook, push_notebook\n",
"from bokeh.plotting import figure, ColumnDataSource\n",
"from bokeh.models import HoverTool\n",
"from bokeh.models.widgets.tables import DataTable, TableColumn\n",
"from bokeh.models import Span\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"output_notebook()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"all_df = df_1.append(df_2, ignore_index=True)\n",
"all_df = all_df.append(df_3, ignore_index=True)\n",
"all_df = all_df.append(df_4, ignore_index=True)\n",
"\n",
"def round_float(s):\n",
" if s.dtype == 'float64':\n",
" s = s.round(7)\n",
" return s\n",
"data = output\n",
"input_json = args\n",
"avg_spend_prop = float(data['spend_axis'])\n",
"\n",
"tooltips=[\n",
" ('lvl 0:', '@level_0'),\n",
" ('lvl 1:', '@level_1'),\n",
" ('lvl 2:', '@level_2'),\n",
" ('lvl 3:', '@level_3'),\n",
" (input_json['goal_metric'], '@'+input_json['goal_metric']),\n",
" ('Projected % cpt change', '@Projected_change'),\n",
" ('Spend Prop', '@spend_prop')\n",
"]\n",
"\n",
"hover = HoverTool(tooltips=tooltips)\n",
"hover1 = HoverTool(tooltips=tooltips)\n",
"\n",
"def build_bokeh_trace(df_obj, num):\n",
" df_obj = df_obj.copy()\n",
" if len(df_obj):\n",
" df_obj = df_obj.sort_values(by='imp_factor', ascending=False)\n",
" df_plotly = df_obj.iloc[0:num].copy()\n",
" df_plotly['idx'] = df_plotly.index\n",
" df_plotly['size'] = df_plotly['imp_factor']\n",
" # range r1 to r2\n",
" r1=8\n",
" r2=30\n",
" min_r = df_plotly['size'].min()\n",
" max_r = df_plotly['size'].max()\n",
" df_plotly['size'] = ((df_plotly['size'] - min_r )*(r2-r1) / (max_r - min_r)) + r1\n",
" df_plotly['overall'] = data['goal_axis']\n",
" df_plotly['overall_spend_prop'] = avg_spend_prop\n",
" else:\n",
" return ColumnDataSource(df_obj)\n",
"\n",
" source = ColumnDataSource(df_plotly)\n",
" return source\n",
"\n",
"spend_source_1 = build_bokeh_trace(df_1, 10)\n",
"spend_source_2 = build_bokeh_trace(df_2, 10)\n",
"spend_source_3 = build_bokeh_trace(df_3, 10)\n",
"spend_source_4 = build_bokeh_trace(df_4, 10)\n",
"if metrics_dict[goal_metric]['diff'] == -1:\n",
" df_2, df_3 = df_3, df_2\n",
" df_4, df_1 = df_1, df_4\n",
"table_source_1 = ColumnDataSource(df_1.apply(round_float))\n",
"table_source_2 = ColumnDataSource(df_2.apply(round_float))\n",
"table_source_3 = ColumnDataSource(df_3.apply(round_float))\n",
"table_source_4 = ColumnDataSource(df_4.apply(round_float))\n",
"\n",
"columns = [\n",
" TableColumn(field=\"level_0\", title=\"level 0\"),\n",
" TableColumn(field=\"level_1\", title=\"level 1\"),\n",
" TableColumn(field=\"level_2\", title=\"level 2\"),\n",
" TableColumn(field=\"level_3\", title=\"level 3\"),\n",
" TableColumn(field=\"other_{0}\".format(input_json['goal_metric']), title=\"Other {0}\".format(input_json['goal_metric'])),\n",
" TableColumn(field=input_json['goal_metric'], title=input_json['goal_metric'],),\n",
" TableColumn(field=\"spend_prop\", title=\"Spend %\"),\n",
" TableColumn(field=\"imp_factor\", title=\"Impact factor\"),\n",
"]\n",
"\n",
"\n",
"# Spend sorted.\n",
"p1 = figure(plot_width=800, plot_height=500, tools=[hover1],\n",
" title=\"Performance by paths.\")\n",
"p1.xaxis.axis_label = 'Spend Proportion'\n",
"p1.yaxis.axis_label = input_json['goal_metric']\n",
"\n",
"if len(df_1):\n",
" prop_plot_1 = p1.circle('spend_prop', input_json['goal_metric'], fill_color='orange', source=spend_source_1, size='size')\n",
"if len(df_2):\n",
" prop_plot_2 = p1.circle('spend_prop', input_json['goal_metric'], fill_color='red', source=spend_source_2, size='size')\n",
"if len(df_3):\n",
" prop_plot_3 = p1.circle('spend_prop', input_json['goal_metric'], fill_color='green', source=spend_source_3, size='size')\n",
"if len(df_4):\n",
" prop_plot_4 = p1.circle('spend_prop', input_json['goal_metric'], fill_color='blue', source=spend_source_4, size='size')\n",
"\n",
"# overall_line = p1.line('spend_prop', 'overall', source=bad_spend_source)\n",
"avg_spend = Span(location=avg_spend_prop, dimension='height', line_color='black', line_width=3)\n",
"avg_cpt = Span(location=data['goal_axis'], dimension='width', line_color='black', line_width=3)\n",
"\n",
"p1.add_layout(avg_spend)\n",
"p1.add_layout(avg_cpt)\n",
"# overall_spend_prop = p1.line('overall_spend_prop', goal_metric, source=bad_spend_source)\n",
"# overall_spend_prop = p1.line('overall_spend_prop', goal_metric, source=good_spend_source)\n",
"\n",
"prop_plot = show(p1, notebook_handle=True)\n",
"\n",
"# Tables.\n",
"print('Q1')\n",
"if len(df_1):\n",
" data_table_1 = DataTable(source=table_source_1, columns=columns, width=1000)\n",
" table_1 = show(data_table_1, notebook_handle=True)\n",
"\n",
"print('Q2')\n",
"if len(df_2):\n",
" data_table_2 = DataTable(source=table_source_2, columns=columns, width=1000)\n",
" table_2 = show(data_table_2, notebook_handle=True)\n",
"\n",
"print('Q3')\n",
"if len(df_3):\n",
" data_table_3 = DataTable(source=table_source_3, columns=columns, width=1000)\n",
" table_3 = show(data_table_3, notebook_handle=True)\n",
"\n",
"print('Q4')\n",
"if len(df_4):\n",
" data_table_4 = DataTable(source=table_source_4, columns=columns, width=1000)\n",
" table_4 = show(data_table_4, notebook_handle=True)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"from jarvis.brain.utility.data_preparation import (get_metrics_date_range, get_channel_metrics_date_range)\n",
"\n",
"def extract_custom_tags(df, custom_tag_column=True):\n",
" \"\"\"Extracting custom tags as individual features.\"\"\"\n",
" if custom_tag_column:\n",
" df = pd.concat([df, pd.DataFrame((d for idx, d in df['custom_tags'].iteritems()))], axis=1)\n",
" del df['custom_tags']\n",
"\n",
" selected_columns = [col for col in df.columns\n",
" if (type(df[col].iloc[0]) != dict or df[col][:1].iloc[0]['source'] in ['AUTO', 'USER'])]\n",
" selected_columns.remove('Campaigns') \n",
" selected_columns.remove('Adsets')\n",
" selected_columns.remove('Ads') \n",
"\n",
" df_ = df[selected_columns]\n",
" df_ = df_.applymap(lambda x: x['name'] if (type(x) == dict and 'name' in x) else x)\n",
" df__ = df[['Campaigns', 'Adsets', 'Ads']]\n",
" df__=df__.applymap(lambda x: x['metadata']['name'] if (type(x) == dict and 'metadata' in x and 'name' in x['metadata']) else x)\n",
" return df_.join(df__)\n",
"\n",
"df = get_metrics_date_range(args['account_ids']['facebook'], 'ad', args['start_date'], args['end_date'],\n",
" metrics + [path_pre_process.get_channel_spend()], dimensions=['custom_tags'], custom_tags=False,\n",
" active=False, rails_api=True, page_size=4000, rename_id=False,\n",
" filters=args['filters'], ad_channel='facebook')\n",
"\n",
"df = extract_custom_tags(df)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"from colorama import Fore, Back, Style\n",
"def path_to_dict(df1,df,good=True):\n",
" ads=[]\n",
" tot_imp=0\n",
" df2=df1.copy()\n",
" for index,row in df2.iterrows():\n",
" key_list=[]\n",
" value_list=()\n",
" df_=df.copy()\n",
" path_dict=dict()\n",
" path_dict[index]=dict()\n",
" for i in range(0, num_levels):\n",
" if(row['level_' +str(i)] != '- : -'):\n",
" level = row['level_' + str(i)]\n",
" level = level.split(':')\n",
" key = level[0]\n",
" key=key.strip()\n",
" level = level[1]\n",
" level = level.strip()\n",
" path_dict[index][key]=level\n",
" for key, value in path_dict[index].items():\n",
" key_list.append(key)\n",
" value_list=value_list+(value ,)\n",
" df_=df_[df_[key]==value]\n",
" df_ = df_.applymap(lambda x: round(x,6) if type(x)==np.float64 else x)\n",
" total_spend = df_.fb_spend.sum()\n",
" total_spend=round(total_spend,5)\n",
" total_num = df_[metrics_dict[goal_metric]['numerator']].sum()\n",
" total_num =round(total_num,5)\n",
" total_deno = df_[metrics_dict[goal_metric]['denominator']].sum()\n",
" total_deno = round(total_deno,5)\n",
" try:\n",
" goal_metric_path=total_num/total_deno\n",
" goal_metric_path=round(goal_metric_path,5)\n",
" except:\n",
" goal_metric_path=0\n",
" df_=df_.groupby(['Campaigns','Adsets','Ads'],as_index=False).agg(aggregations)\n",
" df_=df_.sort('fb_spend',ascending=False)\n",
" df_=df_.reset_index(drop=True)\n",
" df_[goal_metric+' : \\n'+str(goal_metric_path)]=round(df_[metrics_dict[goal_metric]['numerator']]/df_[metrics_dict[goal_metric]['denominator']],5)\n",
" df_.replace([pd.np.inf, -pd.np.inf], 0, inplace=True)\n",
" df_=df_.rename(columns = {metrics_dict[goal_metric]['numerator']:column_names['numerator']+' : \\n'+str(total_num),metrics_dict[goal_metric]['denominator']:column_names['denominator']+' : \\n'+str(total_deno)})\n",
"# df_=df_.drop(['index'], axis = 1, inplace = True)\n",
" print('Path :'+str(index))\n",
" print(\"---------\")\n",
" path_df=df2.loc[[index]]\n",
" \n",
" str_path = Fore.BLUE + path_df['level_0'][index] +'\\n'+ path_df['level_1'][index] + '\\n'+path_df['level_2'][index] #+'\\n'+ path_df['level_3'][index]\n",
" str_path=str_path.replace('- : -','')\n",
" str_path=str_path.strip()\n",
" print(str_path)\n",
" print('\\n')\n",
" print('Spend Proportion here is '+ str(round(row['spend_prop'],5))+' %')\n",
" \n",
" if good:\n",
" bg_color='lightgreen'\n",
" else:\n",
" bg_color='pink'\n",
" df_=df_.style.set_table_styles(\n",
" [{'selector': 'th',\n",
" 'props': [('background-color', bg_color)]}] )\n",
" display(df_)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"column_names=metrics_dict[goal_metric]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"path_to_dict(df_1,df)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"path_to_dict(df_2,df)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"path_to_dict(df_3,df,False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"path_to_dict(df_4,df,False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"anaconda-cloud": {},
"kernelspec": {
"display_name": "Python [jarvis1]",
"language": "python",
"name": "Python [jarvis1]"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.5.2"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment