Ingxaki ephambili enxulumene neekhowudi zezibalo ezisezantsi kwiPython kukuba kunokuba nzima ukuqonda nokutolika iziphumo. I-Python lulwimi olunamandla, kodwa kunokuba nzima ukuyifunda kunye nokuqonda ikhowudi esetyenziselwa izibalo ezingenasiphelo. Ukongeza, kukho iipakethe ezininzi ezahlukeneyo ezikhoyo kwiinkcukacha-manani ezisezantsi kwiPython, enokwenza kube nzima ukukhetha eyona ilungileyo kuhlalutyo oluthile. Okokugqibela, ezinye zezi phakheji zisenokungabi sexesheni okanye zithembeke njengezinye, ngoko ke kubalulekile ukwenza uphando ngaphambi kokuba uzisebenzise.
1. Chi-Square Test of Independence: from scipy.stats import chi2_contingency chi2, p, dof, expected = chi2_contingency(observed) 2. One-Way ANOVA: from scipy import stats F, p = stats.f_oneway(sample1, sample2, sample3) 3. Pearson’s Correlation Coefficient: from scipy.stats import pearsonr corr, _ = pearsonr(x, y)
Umgca 1: Lo mgca ungenisa umsebenzi we-chi2_contingency kwilayibrari ye-scipy.stats, kwaye emva koko uyisebenzise ukubala uvavanyo lwe-chi-square lokuzimela kwidatha ephawulweyo. Iziphumo zolu vavanyo zigcinwa kwi variables chi2, p, dof, kwaye kulindeleke.
Umgca 2: Lo mgca ungenisa umsebenzi we-f_oneway kwilayibrari ye-scipy, kwaye uyisebenzise ukubala indlela enye ye-ANOVA kwiisampuli ezintathu (isampuli1, isampuli2, isampuli3). Iziphumo zolu vavanyo zigcinwe kwi-variables F kunye ne-p.
Umgca 3: Lo mgca ungenisa umsebenzi wepearsonr kwilayibrari ye-scipy.stats, kwaye emva koko uyisebenzise ukubala i-coefficient yonxulumano kaPearson phakathi kwezinto ezimbini eziguquguqukayo (x kunye no-y). Iziphumo zolu vavanyo zigcinwe kwi-variables corr kunye _.
Yintoni izibalo-manani
Ubalo-manani licandelo leenkcukacha-manani elisebenzisa idatha esuka kwisampulu ukwenza intelekelelo okanye ulungelelwaniso malunga nabemi. Ibandakanya ukwenza izigqibo malunga nabemi ngokusekelwe kwidatha eqokelelwe kwisampuli. Kwi-Python, amanani-manani angaphantsi angasetyenziselwa ukwenza izigqibo kunye nokwenza uqikelelo ngokusebenzisa iindlela ezahlukeneyo ezifana novavanyo lwe-hypothesis, uhlalutyo lokulungelelanisa, uhlalutyo lokubuyela umva, kunye nokunye. Ezi ndlela zobuchule zisivumela ukuba sithathe iimbono ezinentsingiselo kwidatha yethu kwaye zisincede senze izigqibo ezingcono.
Iindidi zamanani-nkcazo
KwiPython, kukho iindidi ezininzi zeenkcukacha-manani ezinokuthi zisetyenziswe ukuhlalutya idatha. Ezi ziquka iimvavanyo ze-t, i-ANOVA, iimvavanyo ze-chi-square, iimvavanyo zokulungelelanisa, kunye nohlalutyo lokubuyisela. Iimvavanyo ze-T zisetyenziselwa ukuthelekisa iindlela zamaqela amabini okanye ngaphezulu edatha. I-ANOVA isetyenziselwa ukuthelekisa iindlela zamaqela amaninzi edatha. Iimvavanyo ze-Chi-square zisetyenziselwa ukuvavanya ubudlelwane phakathi kwezinto eziguquguqukayo. Iimvavanyo zonxulumano zilinganisa amandla kunye nesalathiso sobudlelwane bomgca phakathi kwezinto ezimbini eziguquguqukayo. Ekugqibeleni, uhlalutyo lokubuyisela lusetyenziselwa ukuqikelela ukuguquguquka okuxhomekekayo ukusuka kwelinye okanye ngaphezulu kwezinto ezizimeleyo.
Uzibhala njani iinkcukacha-manani
Izibalo-manani licandelo leenkcukacha-manani elisebenzisa idatha esuka kwisampulu ukwenza intelekelelo malunga nabemi apho isampuli ithathwe khona. KwiPython, izibalo ezingeyomfuneko zinokwenziwa kusetyenziswa amathala eencwadi ahlukeneyo anje ngeSciPy, iStatsModels, kunye neNumPy.
Ukwenza izibalo ezisezantsi kwiPython, kuya kufuneka ungenise kuqala amathala eencwadi ayimfuneko kwaye emva koko usebenzise imisebenzi enje nge mean(), median(), mode(), umahluko(), utembuko olusemgangathweni(), t-test(), chi -square test() etc. Umzekelo, ukuba ubufuna ukubala intsingiselo yesethi yedatha enikiweyo, ungasebenzisa i mean() umsebenzi ukusuka kwiNumPy:
ngenisa i-numpy njenge-np
idatha = [1,2,3,4]
mean_value = np.mean(data)
shicilela(ixabiso_elithetha) # Isiphumo: 2.5
