Semoss/py/analytics.py at dev · SEMOSS/Semoss · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
import string
import pandas as pd
import random
import numpy as np
from sklearn.linear_model import LinearRegression


class PyFrame:

    def __init__(self, cache):
        self.cache = cache

    @classmethod
    def makefm(cls, frame):
        cache = {}
        cache["data"] = frame
        cache["version"] = 0
        cache["low_version"] = 0
        cache[0] = frame
        return cls(cache)

    @classmethod
    def makefm_csv(cls, fileName):
        frame = pd.read_csv(fileName)
        cache = {}
        cache["data"] = frame
        cache["version"] = 0
        cache["low_version"] = 0
        cache[0] = frame
        return cls(cache)

    def id_generator(self, size=6, chars=string.ascii_uppercase + string.digits):
        return "".join(random.choice(chars) for _ in range(size))

    def makeCopy(this):
        version = 0
        if version in this.cache:
            if "high_version" in this.cache:
                version = this.cache["high_version"]
            else:
                version = this.cache["version"]
            version = version + 1

        this.cache[version] = this.cache["data"].copy()
        this.cache["version"] = version
        this.cache["high_version"] = version

    # falls back to the last version
    def fallback(this):
        version = this.cache["version"]
        low_version = this.cache["low_version"]
        if version in this.cache and version > low_version:
            this.cache["high_version"] = version
            version = version - 1
            this.cache["version"] = version
            this.cache["data"] = this.cache[version]
        else:
            print("cant fall back. In the latest")

    def get_correlation(this, columns="assign", inplace=True):
        print(columns)
        output = "to be assigned"
        if columns == "assign":
            output = this.cache["data"].corr().unstack().reset_index()
        else:
            print("in the else")
            output = this.cache["data"][columns].corr().unstack().reset_index()
        output.columns = ["Column_Header_X", "Column_Header_Y", "Correlation"]
        return output

    def get_regression(this, target, source):
        model = LinearRegression(fit_intercept=True)
        frame = this.cache["data"]
        target = frame[target]
        if len(source) == 1:
            # print ('In Single Source Column')
            temp = source[0]
            # print('Recasting', temp)
            source = frame[temp][:, np.newaxis]
        else:
            source = frame[source]
        model.fit(source, target)
        predict = list(model.predict(source))
        row_id = list(range(0, len(predict)))
        target = list(target)
        # print(len(row_id), len(target), len(predict))
        result = pd.DataFrame({"row": row_id, "actual": target, "fitted": predict})
        result.columns = ["ROW_ID", "Actual", "Fitted"]
        return result