datawhispers package

Submodules

datawhispers.advancedProg module

datawhispers.advancedProg.PolyCoefficients(x, coeffs)

Returns a polynomial for x values for the coeffs provided.

The coefficients must be in ascending order (x**0 to x**o).

class datawhispers.advancedProg.Trend(x, y, ansatz, deg=None)

Bases: object

Trends Class. Trend objects have values and method attributes.

coef()

Computes coefficients of corresponding ansatz

make_easy_plot(file_name)

Shows a plot of the data, the regression and saves the plot

pred(x)

Computes the predction for input x and the computed corresponding coefficients

r2()

Computes the coefficient of determination for the training input

datawhispers.advancedProg.add_mnist_num_arrays(num1, num2)

Returns the image of the result and saves it as mnist_result.png

Parameters:

• num1 – np.array of length (784,) or (28,28)

• num2 – np.array of length (784,) or (28,28)

Returns:

Outputs the image

Raises:

None –

datawhispers.advancedProg.expReg(x_in, y)

Time series exponential regression.

datawhispers.advancedProg.freeReg(x_in, y_out, ansatz)

Regression with user ansatz. The ansatz is expected to depend on three parameters, a, b, and c. The ansatz is expected to be a string with a symbolic formulation. for instance: ‘a*arctan(b*x_in+c)’.

Parameters:

• x_in – Array with x-values

• y – Array with y-values

• ansatz –

  “linReg”, “polReg”, “trigReg” or “expReg”

  Returns:

  coefs

  Raises:

  None

datawhispers.advancedProg.leastSquares(func, x)

Solve a nonlinear least-squares problem with bounds on the variables.

Parameters:

• func – Function which computes the vector of residuals,

• fun (with the signature) –

• shape (It must allocate and return a 1-D array_like of) –

• shape –

• residuals (If the argument x is complex or the function fun returns complex) –

:param : :param it must be wrapped in a real function of real arguments: :param as shown at the end of the Examples section.: :param x: Array with x-values

Returns:

ndarray, shape (n,)

Solution found.

cost: float

Value of the cost function at the solution.

func: ndarray, shape (m,)

Vector of residuals at the solution.

Return type:

x

Raises:

None –

datawhispers.advancedProg.linReg(x_in, y)

Time series linear regression. Returns coefs in polynomial descending order. Coefs computed analytically.

Args:

x_in: Array with x-values y: Array with y-values

Returns:

coefs in descending order

Raises:

None

datawhispers.advancedProg.make_plot(x, y, y_reg, xticks=[], yticks=[], xlabel='x', ylabel='y', colors=['lightblue', 'black'], name='fig_reg.png')

Outputs a graph for (x and y) and (x and y_reg) and saves it as fig_reg.png

Parameters:

• x – array with x-values

• y – array with y-values

• y_reg – array with regression y_values

• xticks (optional) – list with values to use as x-ticks

• yticks (optional) – list with values to use as y-ticks

• xlabel (optional) – defualt “x”

• ylabel (optional) – defualt “y”

• colors (optional) – default [“lightblue”, “black”] scatter=lightblue and line=black

• name (optional) – default “fig_reg.png”

Returns:

Outputs the graph and saves it

Raises:

None –

datawhispers.advancedProg.plot_all_regs(x, y, xticks=None, yticks=None)

Returns the regression of all types and saves them as png

Parameters:

• x – array with x-values

• y – array with y-values

• xticks (optional) – list with values to use as x-ticks

• yticks (optional) – list with values to use as y-ticks

Returns:

Outputs the graphs and saves them

Raises:

None –

datawhispers.advancedProg.polReg(x_in, y, deg)

Time series polynomial regression. Returns coefs in polynomial descending order. Coefs computed numerically.

Parameters:

• x_in – Array with x-values

• y – Array with y-values

• deg –

  the degree of the polynomial

  Returns:

  coefs in descending order

  Raises:

  None

datawhispers.advancedProg.predict(ansatz, coef, x_in, freeRegAnsatz=None)

Computes the predction for input x_in and the computed corresponding coefficients

datawhispers.advancedProg.r2(y, y_pred)

Coefficient of determination

datawhispers.advancedProg.show_mnist_from_array(arr)

Returns the image of the mnist number and saves it as mnist_num.png

Parameters:

arr – of size (784,) or (28,28) with values from 0 to 255

Returns:

Outputs the image

Raises:

None –

datawhispers.advancedProg.show_mnist_from_file(filepath)

Returns the images of the mnist numbers in the file

Parameters:

filepath – csv-filepath with lines consisting of values from 0 to 255 with length of 785 or 784

Returns:

Outputs the images

Raises:

None –

datawhispers.advancedProg.trigReg(x_in, y)

Time seriessine regression. Returns amplitude, frequency and phase

Parameters:

• x_in – Array with x-values

• y – Array with y-values

Returns:

amplitude, frequency and phase

Raises:

None –

datawhispers.datavis module

datawhispers.datavis.Analyse_Tabelle(df: DataFrame, Zielvariable: str, Anzahl_an_gewünschten_Variablen: Optional[int] = None, Zielvariable_good='', Zielvariable_bad='', Zielvariable_Grenze: int = 0, kategorische_Variablen=[], Numerische_Variablen=[])

Gibt die Tabelle mit den abhängigsten Variablen aus. @param df: Der DataFrame @param Zielvariable: Zielvariable @param Anzahl_an_gewünschten_Variablen (opt.) @param Zielvariable_good: Nur bei KATEGORISCHER Zielvariable @param Zielvariable_bad: Nur bei KATEGORISCHER Zielvariable @param Zielvariable_Grenze: Nur bei NUMERISCHER Zielvariable ;Schwellwert ab wann Gut oder Schlecht sonst Durchschnitt @param kategorische_Variablen: Eingeben einer Liste an kategorischen Variablen, die im Datensatz als numerisch gekennzeichnet sind (opt.) @param numerische_Variablen: Eingeben einer Liste an numerischen Variablen, die im Datensatz als kategorisch gekennzeichnet sind (opt.) @return Tabelle mit Variablen, Kennwerten, Testverfahren

datawhispers.datavis.Cor_tab(df, max=10)

class datawhispers.datavis.DataVisAnalyse(df, classification_column, good, bad)

Bases: object

get_T_Test(df=None)

[metrisch] - [kategorisch]

Stellt die unterschiedlichen stat. Abhängigkeiten mit Mittelwert und Standardabweichung visuell dar @self.classification_column: Nur zwei Ausprägungen @return [Variable,p-Wert,t-wert,mü-gut, sigma-gut, mü-schlecht, sigma-schlecht]

get_all()

Eine rundumanalyse eines Datensatzes in statistischer Form

get_chi2(varname, df=None)

[kategorisch] - [kategorisch]

get_corr_table(max: int = 10, df=None)

[numerisch] - [numerisch]

Stellt die stärksten Korrelationen zwischen intervallskalierten Variablen sortiert visuell dar.

Größeres p –> Größere Abhängigkeit @Variabel df: Zugehöriges DataFrame @Variabel max: Maximale Anzahl der Korrelationen

get_most_relevant(ignore, n=5)

Get a df with most relevant variables

Parameters:

• ignore (list) – list with column_names which to ignore

• n (int) – features to inspect

get_prcnt_diff(val1, val2)

nice()

datawhispers.datavis.Datensatz_bereinigen(df: DataFrame, kategorische_Spalten: list = [], numerische_Spalten: list = [])

Wandelt ‘Numerische Daten’ in Kategorische und anders herum. @return Bereinigter Datensatz als pd.DataFrame

datawhispers.datavis.T_Test_Rückgabe(df: DataFrame, Zielvariable, Zielvariable_Wert_gut, Zielvariable_Wert_schlecht, Numerische_Grenze, Grafik=False)

[metrisch] - [kategorisch]

Stellt die unterschiedlichen stat. Abhängigkeiten mit Mittelwert und Standardabweichung visuell dar @Zielvariable: Nur zwei Ausprägungen @return [Variable,p-Wert,t-wert,mü-gut, sigma-gut, mü-schlecht, sigma-schlecht]

datawhispers.datavis.chi_square_all_Rückgabe(df: DataFrame, Zielvariable, Numerische_Grenze)

[kategorisch] - [kategorisch]

Erstellt alle möglichen chi2-Werte und p werte

p<0.05 und chi2 sehr groß –> Signifikant

datawhispers.datavis.chi_square_einzeln_Rückgabe(df: DataFrame, varname, Zielvariable, Numerische_Grenze=0)

[kategorisch] - [kategorisch]

datawhispers.datavis.make_scatter_plot(x, y, width: int = 8, height: int = 6, xlabel: str = '', ylabel: str = '', title: str = '', color='gray', filename: str = 'fig.png')

Outputs a scatter plot

Parameters:

• x – array with x-values

• y – array with y-values

• xticks (optional) – list with values to use as x-ticks

• yticks (optional) – list with values to use as y-ticks

• xlabel (optional) – str with xlabel

• ylabel (optional) – str with ylabel

• title (optional) – str with title

• color (optional) – str or tuple consisting of rgb values for color

• filename (optional) – str with filename or path + filename

Returns:

Outputs the graph and saves it

Raises:

None –

datawhispers.mathFuncs module

Module contents