In [ ]:

    

import ClassiferHelperAPI as CH
import importlib
import numpy as np
import pandas as pd
importlib.reload(CH)
from ast import literal_eval
import plotly.plotly as py
import htmltag as HT
import cufflinks as cf # this is necessary to link pandas to plotly
cf.go_offline()
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
from collections import Counter
import csv
import plotly.graph_objs as go


    

In [ ]:

    

def plot(rgrObj,arr,arr_min,title,flNm,range,errorBar = True):
    trace1 = go.Scatter(
        x = list(rgrObj.preds),
        y = list(rgrObj.test_y),
        error_y = dict(
            type='data',
            symmetric = False,
            array = arr,
            arrayminus = arr_min,
            visible=errorBar
        ),
        mode = 'markers'
    )

    layout= go.Layout(
        title= title,
        xaxis= dict(
            title= 'Predicted Share Rate',
            ticklen= 5,
            zeroline= False,
            gridwidth= 2,
            range=[-5,110]
        ),
        yaxis=dict(
            title= 'Actual Share rate',
            ticklen= 5,
            gridwidth= 2,
            #range=range
        )
    )
    
    data = [trace1]

    fig = dict(data=data,layout=layout)
    a = py.iplot(fig,filename=flNm)
    
    return a

def runRgr(methodName,attribType):
    if attribType = "beauty":
        inpData = pd.DataFrame.from_csv("../data/BeautyFtrVector_GZC.csv")
        inpData.reindex(np.random.permutation(inpData.index))
        y = inpData['Proportion']
        inpData.drop(['Proportion'],1,inplace=True)
        train_x, test_x, train_y, test_y = CH.train_test_split(inpData, y, test_size = 0.4)
        rgr = CH.getLearningAlgo(methodName,{'fit_intercept':True})
    
        rgrObj = RgrCls.RegressionCapsule(rgr,methodName,0.8,train_x,train_y,test_x,test_y)
        
    else:
        train_data_fl = "../FinalResults/ImgShrRnkListWithTags.csv"
        infoGainFl = "../data/infoGainsExpt2.csv"
        allAttribs = CH.genAllAttribs(train_data_fl,attribType,infoGainFl)

        rgrObj = CH.buildRegrMod(train_data_fl,allAttribs,0.6,methodName,kwargs={'fit_intercept':True})
    
    rgrObj.runClf(computeMetrics=True,removeOutliers=True)
    
    x = [i for i in range(len(rgrObj.preds))]
    errors = [list(rgrObj.test_y)[i] - list(rgrObj.preds)[i] for i in range(len(rgrObj.preds))]
    arr = [-1 * errors[i] if errors[i] < 0 else 0 for i in range(len(errors)) ]
    arr_min = [errors[i] if errors[i] > 0 else 0 for i in range(len(errors)) ]
    return rgrObj,arr,arr_min


    

In [ ]:

    

attribTypes = ['sparse','non_sparse','non_zero','abv_mean']
rgrAlgoTypes = ['linear','ridge','lasso']

embedCodes = []
for attrib in attribTypes:
    code = []
    for alg in rgrAlgoTypes:
        rgrObj, arr,arr_min = runRgr(alg,attrib)
        title = "%s regression results using %s attributes" %(alg,attrib)
        flNm = "%s_regession_%s_attributes_%s" %(alg,attrib,str(True))
        a = plot(rgrObj,arr,arr_min,title,flNm,[-100,200],errorBar = True)
        code.append(a.embed_code)    
    embedCodes.append(code)


    

In [ ]:

    

attribTypes = ['sparse','non_sparse','non_zero','abv_mean']
rgrAlgoTypes = ['linear','ridge','lasso']

embedCodes = []
for attrib in attribTypes:
    code = []
    for alg in rgrAlgoTypes:
        rgrObj, _, _ = runRgr(alg,attrib)
        title = "%s regression results using %s attributes" %(alg,attrib)
        flNm = "%s_regession_%s_attributes_%s" %(alg,attrib,str(False))
        a = plot(rgrObj,errs,[],title,flNm,[-10,110],errorBar = False)
        code.append(a.embed_code)
    embedCodes.append(code)


    

In [ ]:

    

for c in embedCodes[3]:
    print(c)


    

In [ ]:

    

rgrObj, arr,arr_min = runRgr('dtree_regressor','non_sparse')


    

In [ ]:

    

title = "%s regression results using %s attributes" %(alg,attrib)
flNm = "%s_regession_%s_attributes_%s" %(alg,attrib,str(True))
a = plot(rgrObj,arr,arr_min,title,flNm,[-100,200],errorBar = True)


    

In [ ]:

    

attribTypes = ['sparse','non_sparse','non_zero','abv_mean']
rgrAlgoTypes = ['linear_svr','svr','dtree_regressor']

embedCodes = []
for attrib in attribTypes:
    code = []
    for alg in rgrAlgoTypes:
        rgrObj, _, _ = runRgr(alg,attrib)
        print("Absolute error for %s using %s : %f" %(alg,attrib,rgrObj.abserr))
        print("Mean Squared error for %s using %s : %f" %(alg,attrib,rgrObj.sqerr))
        title = "%s regression results using %s attributes" %(alg,attrib)
        flNm = "%s_regession_%s_attributes_%s" %(alg,attrib,str(False))
        a = plot(rgrObj,[],[],title,flNm,[-10,110],errorBar = False)
        code.append(a.embed_code)
    embedCodes.append(code)


    

In [ ]:

    

len(embedCodes)


    

In [ ]:

    

attribTypes = ['sparse','non_sparse','non_zero','abv_mean']
rgrAlgoTypes = ['linear_svr','svr','dtree_regressor']

embedCodes = []
for attrib in attribTypes:
    code = []
    for alg in rgrAlgoTypes:
        rgrObj, arr,arr_min = runRgr(alg,attrib)
        title = "%s regression results using %s attributes" %(alg,attrib)
        flNm = "%s_regession_%s_attributes_%s" %(alg,attrib,str(True))
        a = plot(rgrObj,arr,arr_min,title,flNm,[-100,200],errorBar = True)
        code.append(a.embed_code)    
    embedCodes.append(code)


    

In [4]:

    

import PopulationEstimatorFromClf as PE
import importlib
importlib.reload(PE)
import ClassiferHelperAPI as CH


    

In [ ]:

    

train_data_fl = "../FinalResults/ImgShrRnkListWithTags.csv"
test_data_fl = "../data/full_gid_aid_ftr_agg.csv"
methodName = "linear"
attribType = "non_sparse"
infoGainFl=None
regrArgs=kwargsDict
obj, pred_results = PE.trainTestRgrs(train_data_fl,
                test_data_fl,
                methodName,
                attribType,
                 None,
                regrArgs=kwargsDict)


    

In [ ]: