adhadse

from sklearn.model_selection import cross_val_predict
from sklearn.metrics import precision_recall_curve

# First you need to get the precisions, recalls, thresholds for your classifier.
y_scores = cross_val_predict(classifier, X_train, y_train, cv=3, method='decision_function')
precisions, recalls, thresholds = precision_recall_curve(y_train, y_scores)

adhadse

def plot_precision_vs_recall(precisions, recalls, metric_name=None, metric_perc=None):
    plt.figure(figsize=(15, 10))
    plt.plot(recalls, precisions, 'b-', linewidth=2)
    plt.xlabel("Recall", fontsize=15)
    plt.ylabel("Precision", fontsize=15)
    plt.axis([0, 1, 0, 1])
    if metric_name=='precision':
        # tradedoff recall  & threshold at percentage precision we want.
        recall_atperc_precision = recalls[np.argmax(precisions >= metric_perc)] 
        threshold_atperc_precision = thresholds[np.argmax(precisions >= metric_perc)]

adhadse

In[1]  y_scores = clf.decision_function([instance_from_test_set])
       no_threshold = 0
       y_some_prediction = (y_scores > no_hreshold)
       y_some_prediction
Out[1] array([True])

In[2]  # Threshold returned from plotting func
       y_some_prediction = (y_score > threshold) 
       y_some_prediction
Out[2] array([False])

adhadse

def determine_outlier_thresholds_iqr(dataframe, col_name, th1=0.25, th3=0.75):
    quartile1 = dataframe[col_name].quantile(th1)
    quartile3 = dataframe[col_name].quantile(th3)
    iqr = quartile3 - quartile1
    upper_limit = quartile3 + 1.5 * iqr
    lower_limit = quartile1 - 1.5 * iqr
    return lower_limit, upper_limit

adhadse

def check_outliers_iqr(dataframe, col_name):
    lower_limit, upper_limit = determine_outlier_thresholds_iqr(dataframe, col_name)
    if dataframe[(dataframe[col_name] > upper_limit) | (dataframe[col_name] < lower_limit)].any(axis=None):
        return True
    else: 
        return False

def replace_with_thresholds_iqr(dataframe,cols, th1=0.05, th3=0.95, replace=False):
    from tabulate import tabulate
    data = []

adhadse

def determine_outlier_thresholds_std(dataframe, col_name):
    upper_boundary = dataframe[col_name].mean() + 3 * dataframe[col_name].std()
    lower_boundary = dataframe[col_name].mean() - 3 * dataframe[col_name].std()
    return lower_boundary, upper_boundary

adhadse

def check_outliers_std(dataframe, col_name):
    lower_boundary, upper_boundary = determine_outlier_thresholds_std(dataframe, col_name)
    if dataframe[(dataframe[col_name] > upper_boundary) | (dataframe[col_name] < lower_boundary)].any(axis=None):
        return True
    else: 
        return False

def replace_with_thresholds_std(dataframe, cols, replace=False):
    from tabulate import tabulate
    data = []

adhadse

import tensorflow as tf
from tensorflow import keras

adhadse

model = keras.models.Sequential()
model.add(keras.layers.Flatten(input_shape=[28, 28]))
model.add(keras.layers.Dense(300, activation="relu"))
model.add(keras.layers.Dense(100, activation="relu"))
model.add(keras.layers.Dense(10, activation="softmax"))

adhadse

model = keras.models.Sequential([
    keras.layers.Flatten(input_shape=[28, 28]),
    keras.layers.Dense(300, activation="relu"),
    keras.layers.Dense(100, activation="relu"),
    keras.layers.Dense(10, activation="softmax")
])