示例#1
0
import pandas
import po
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('filename', type=str)
args = parser.parse_args()

df = po.read_csv(args.filename, sep="\t")


correctly_mapped = df.query('(indel_pos - mapped_pos == 21)')
correctly_aligned = pandas.DataFrame([ r for r in correctly_mapped.values if r[2].startswith('22') ], columns=df.keys())
print("\t\ttotal\tmap correctly\tsame with profile")
print(args.filename,"\t",len(df),"\t",len(correctly_mapped),"\t",len(correctly_aligned))

if len(correctly_aligned) > 0:
	correctly_aligned_by_chr = correctly_aligned[['ChrID','non_gap_num','opt_num_1_alignment']].groupby('ChrID', sort=False)
	print(correctly_aligned_by_chr.mean())


## correctly mapped and number of optimal affine-gapped alignment is greater than 1.
correctly_mapped = df.query('(indel_pos - mapped_pos == 21) and (opt_num_1_alignment > 1)')
correctly_aligned = pandas.DataFrame([ r for r in correctly_mapped.values if r[2].startswith('22') ], columns=df.keys())
print("correctly mapped, same w profile and >1 multiple opt alignments")
print(len(correctly_aligned))
if len(correctly_aligned) > 0:
	correctly_aligned_by_chr = correctly_aligned[['ChrID','non_gap_num','opt_num_1_alignment']].groupby('ChrID', sort=False)
	print(correctly_aligned_by_chr.mean())
示例#2
0
import po

# get data from machineuft predict use LogisticRegression and save it in L_machine
iris = po.read_csv("data\\F2011-F2015_Regression_FulColumn.csv", encoding='utf-8')

iris.Classify(["ARESD", "ACT"], "REGTYPE", method="LogisticRegression", portion=0.8)

iris.to_csv("Registration_All_Logistic.csv", encoding='utf-8')


# get data from machineuft predict use RandomForest and save it in L_machine
iris = po.read_csv("data\\F2011-F2015_Regression_FulColumn.csv", encoding='utf-8')

iris.Classify(["ARESD", "ACT"], "REGTYPE", method="RandomForest", portion=0.9)

iris.to_csv("Registration_All_RandomForest.csv", encoding='utf-8')



# get data from machineuft predict use Naive-Bayes and save it in L_machine
iris = po.read_csv("D:\\MasterProject\\po-master\\po\\data\\F2011-F2015_Regression_FulColumn.csv", encoding='utf-8')

iris.Classify(["ARESD", "ACT", "HSGPA", "STATE", "RACE", "FIRSTGEN", "FATHEREDUCATION", "MOTHEREDUCATION"], "REGTYPE", method="GaussianNB", portion=0.9)

iris.to_csv("GaussianNB_Registration.csv", encoding='utf-8')



# get data from machineuft predict use SVM and save it in L_machine
iris = po.read_csv("D:\\MasterProject\\po-master\\po\\data\\F2011-F2015_Regression_FulColumn.csv", encoding='utf-8')
示例#3
0
import po

# get data from machineuft predict use LinearRegression and save it in L_machine
iris = po.read_csv("data\\iris.csv", encoding="utf-8")

iris.Classify(["SepalLength", "SepalWidth", "PetalLength", "PetalWidth"], "Species", method="RandomForest", portion=0.7)

iris.to_csv("Iris_RandomForest.csv", encoding="utf-8")


# get data from machineuft predict use LinearRegression and save it in L_machine
iris = po.read_csv("data\\iris.csv", encoding="utf-8")

iris.Classify(["SepalLength", "SepalWidth", "PetalLength", "PetalWidth"], "Species", method="SVM", portion=0.7)

iris.to_csv("Iris_SVM.csv", encoding="utf-8")


# get data from machineuft predict use LinearRegression and save it in L_machine
iris = po.read_csv("data\\iris.csv", encoding="utf-8")

iris.Classify(
    ["SepalLength", "SepalWidth", "PetalLength", "PetalWidth"],
    "Species",
    method="KNeighbors",
    n_neighbors=4,
    portion=0.7,
)

iris.to_csv("Iris_Kneighbors.csv", encoding="utf-8")
示例#4
0
import po

# get data from machineuft predict use LinearRegression and save it in L_machine
comp = po.read_csv("data\\computer_hardware.csv", encoding='utf-8')

comp.Regression(["MYCT", "MMIN", "CACH", "CHMIN", "CHMAX", "PRP"], "ERP", method="LinearRegression", portion=0.8)

comp.to_csv("Hardware_Linear.csv", encoding='utf-8')


# get data from machineuft predict use LinearRegression and save it in L_machine
comp = po.read_csv("data\\computer_hardware.csv", encoding='utf-8')

comp.Regression(["MYCT", "MMIN", "CACH", "CHMIN", "CHMAX", "PRP"], "ERP", method="RandomForest", portion=0.8)

comp.to_csv("Hardware_RandomForest.csv", encoding='utf-8')


# get data from machineuft predict use LinearRegression and save it in L_machine
comp = po.read_csv("data\\computer_hardware.csv", encoding='utf-8')

comp.Regression(["MYCT", "MMIN", "CACH", "CHMIN", "CHMAX", "PRP"], "ERP", method="KNeighbors", n_neighbors=4, portion=0.8)

comp.to_csv("Hardware_KN.csv", encoding='utf-8')
示例#5
0
import po
import warnings

warnings.filterwarnings("ignore", category=DeprecationWarning) 


# get data from machineuft predict use RandomForest and save it in L_machine
iris = po.read_csv("data\\F2011-F2015_Regression_FulColumn.csv", encoding='utf-8')

iris.Regression(["ACT"], "FATHEREDUCATION", method="KNeighbors", portion=0.8)
示例#6
0
#Optimal Alignment
import pandas
import po
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('filename', type=str)
args = parser.parse_args()

df = po.read_csv(args.filename, sep="\t")
df["possiblity"] = 1.0/df["opt_num_1_alignment"]

correctly_mapped = df.query('indel_pos - mapped_pos == 21')
tmp = [ r for r in correctly_mapped.values if r[2].startswith('22') and r[6] > 1 ]

correctly_aligned = pandas.DataFrame(tmp, columns=df.keys())
correctly_aligned["p"] = 1.0/correctly_aligned["opt_num_1_alignment"]
correctly_aligned_by_chr = correctly_aligned[['ChrID','non_gap_num','opt_num_1_alignment', 'possiblity']].groupby('ChrID', sort=False)

print("\t\ttotal\tmapped\taligned\taveragepossible")
print(args.filename,"\t",len(df),"\t",len(correctly_mapped),"\t",len(correctly_aligned), correctly_aligned["possiblity"].mean())

print("\nMean")
print(correctly_aligned_by_chr.mean())

print("\nStandard Dev")
print(correctly_aligned_by_chr.std())
Status API Training Shop Blog About