core

Fill in a module description here

Create an instance of the class CalibrationModel by passing in the path to both your calibration data and your sample data. Next specify the name of your response variable and the nummber of test replicates you measured.

/opt/hostedtoolcache/Python/3.9.17/x64/lib/python3.9/site-packages/fastcore/docscrape.py:225: UserWarning: Unknown section Attributes
  else: warn(msg)
/opt/hostedtoolcache/Python/3.9.17/x64/lib/python3.9/site-packages/fastcore/docscrape.py:225: UserWarning: Unknown section Methods
  else: warn(msg)

---

source

CalibrationModel

 CalibrationModel (x, y)

A class used to create a calibration curve for a given set of data.

…

Tests

def generate_test_data(slope, intercept):
        x = np.linspace(1, 10, num=5)
        y = intercept + x * slope
        df = pd.DataFrame({'concentration': x, "abs": y})
        return df
def generate_sample_data():
    x = np.array(['unknown1', 'unknown2'])
    y = np.array([13.75, 20.50])
    df = pd.DataFrame({'sample': x, "abs": y})
    df = df.set_index('sample')
    return df


# test_data = generate_test_data(3, 4)
# sample_data = generate_sample_data()

test_data = pd.DataFrame({'concentration': [0.2, 0.05, 0.1, 0.8, 0.6, 0.4], "abs": [0.221, 0.057, 0.119, 0.73, 0.599, 0.383]})
sample_data = pd.DataFrame({'unknown': [0.490, 0.471, 0.484, 0.473, 0.479, 0.492]})

Create a calibration model object by passing in the predictor and response data for your calibration curve.

cal = CalibrationModel(x=test_data['concentration'], y=test_data['abs'])

Call the fit_model method and the parameters of the fit will be printed to the screen.

cal.fit_model()

Calibration curve
R2 = 0.9976282521058687
Slope = 0.9044109330819979
Intercept = 0.027419415645617395

The individual parameters can be accessed by selecting the parameter name with dot notation. For example, to access the slope of the calibration curve, use the following code:

cal.slope

print(f"Slope: {cal.slope}")

Slope: 0.9044109330819979

Calling the method .linest_stats() will produce a pandas dataframe of the same statistical output you might expect when performing a linest analysis using excel or sheets.

cal.linest_stats()

	Slope	Intercept	Uncertainty in slope	Uncertainty in intercept	Standard error of regression	F-statistic	Degrees of freedom	Regression sum of squares	Residual sum of squares
0	0.904411	0.027419	0.0312	0.017178	0.020745	840.261133	4	0.361606	0.001721

Inverse predictions can be called via the .inverse_prediction method. Call the method and pass a list of values you would like to predict. Note, even if you only have a single sample still pass the sample in a list.

cal.inverse_prediction(sample_data['unknown'])

'0.5020733029033536 ± 0.031053583676141718'

The .calplot method will plot the calibration curve.

cal.calplot(xlab='predictor', ylab='response')