ML-Based Water Portability Prediction

CONTEXT

Access to safe drinking-water is essential to health, a basic human right and a component of effective policy for health protection. This is important as a health and development issue at a national, regional and local level. In some regions, it has been shown that investments in water supply and sanitation can yield a net economic benefit, since the reductions in adverse health effects and health care costs outweigh the costs of undertaking the interventions.

dataframed = pd.read_csv('/kaggle/input/water-potability/water_potability.csv');
dataframed.head()

pH	Hardness	Solids	Chloramines	Sulfate	Conductivity	Organic_carbon	Trihalomethanes	Turbidity
NaN	204.890455	20791.318981	7.300212	368.516441	564.308654	10.379783	86.990970	2.963135
3.716080	129.422921	18630.057858	6.635246	NaN	592.885359	15.180013	56.329076	4.500656
8.099124	224.236259	19909.541732	9.275884	NaN	418.606213	16.868637	66.420093	3.055934
8.316766	214.373394	22018.417441	8.059332	356.886136	363.266516	18.436524	100.341674	4.628771
9.092223	181.101509	17978.986339	6.546600	310.135738	398.410813	11.558279	31.997993	4.075075

The Dataset consists of the following :

3276 rows
10 columns

print('Shape of the Dataset =', dataframed.shape)

Shape of the Dataset = (3276, 10)

METADATA

1. pH value: PH is an important parameter in evaluating the acid–base balance of water. It is also the indicator of acidic or alkaline condition of water status. WHO has recommended maximum permissible limit of pH from 6.5 to 8.5. The current investigation ranges were 6.52–6.83 which are in the range of WHO standards.

2. Hardness: Hardness is mainly caused by calcium and magnesium salts. These salts are dissolved from geologic deposits through which water travels. The length of time water is in contact with hardness producing material helps determine how much hardness there is in raw water. Hardness was originally defined as the capacity of water to precipitate soap caused by Calcium and Magnesium.

3. Solids (Total dissolved solids – TDS): Water has the ability to dissolve a wide range of inorganic and some organic minerals or salts such as potassium, calcium, sodium, bicarbonates, chlorides, magnesium, sulfates etc. These minerals produced un-wanted taste and diluted color in appearance of water. This is the important parameter for the use of water. The water with high TDS value indicates that water is highly mineralized. Desirable limit for TDS is 500 mg/l and maximum limit is 1000 mg/l which prescribed for drinking purpose.

4. Chloramines: Chlorine and chloramine are the major disinfectants used in public water systems. Chloramines are most commonly formed when ammonia is added to chlorine to treat drinking water. Chlorine levels up to 4 milligrams per liter (mg/L or 4 parts per million (ppm)) are considered safe in drinking water.

5. Sulfate: Sulfates are naturally occurring substances that are found in minerals, soil, and rocks. They are present in ambient air, groundwater, plants, and food. The principal commercial use of sulfate is in the chemical industry. Sulfate concentration in seawater is about 2,700 milligrams per liter (mg/L). It ranges from 3 to 30 mg/L in most freshwater supplies, although much higher concentrations (1000 mg/L) are found in some geographic locations.

6. Conductivity: Pure water is not a good conductor of electric current rather’s a good insulator. Increase in ions concentration enhances the electrical conductivity of water. Generally, the amount of dissolved solids in water determines the electrical conductivity. Electrical conductivity (EC) actually measures the ionic process of a solution that enables it to transmit current. According to WHO standards, EC value should not exceeded 400 μS/cm.

7. Organic_carbon: Total Organic Carbon (TOC) in source waters comes from decaying natural organic matter (NOM) as well as synthetic sources. TOC is a measure of the total amount of carbon in organic compounds in pure water. According to US EPA < 2 mg/L as TOC in treated / drinking water, and < 4 mg/Lit in source water which is use for treatment.

8. Trihalomethanes: THMs are chemicals which may be found in water treated with chlorine. The concentration of THMs in drinking water varies according to the level of organic material in the water, the amount of chlorine required to treat the water, and the temperature of the water that is being treated. THM levels up to 80 ppm is considered safe in drinking water.

9. Turbidity: The turbidity of water depends on the quantity of solid matter present in the suspended state. It is a measure of light emitting properties of water and the test is used to indicate the quality of waste discharge with respect to colloidal matter. The mean turbidity value obtained for Wondo Genet Campus (0.98 NTU) is lower than the WHO recommended value of 5.00 NTU.

10. Potability: Indicates if water is safe for human consumption where 1 means Potable and 0 means Not potable.

dataframed.columns.values.tolist()

['ph',
 'Hardness',
 'Solids',
 'Chloramines',
 'Sulfate',
 'Conductivity',
 'Organic_carbon',
 'Trihalomethanes',
 'Turbidity',
 'Potability']

dataframed.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3276 entries, 0 to 3275
Data columns (total 10 columns):
 #   Column           Non-Null Count  Dtype  
---  ------           --------------  -----  
 0   ph               2785 non-null   float64
 1   Hardness         3276 non-null   float64
 2   Solids           3276 non-null   float64
 3   Chloramines      3276 non-null   float64
 4   Sulfate          2495 non-null   float64
 5   Conductivity     3276 non-null   float64
 6   Organic_carbon   3276 non-null   float64
 7   Trihalomethanes  3114 non-null   float64
 8   Turbidity        3276 non-null   float64
 9   Potability       3276 non-null   int64  
dtypes: float64(9), int64(1)
memory usage: 256.1 KB

dataframed.describe()

	ph	Hardness	Solids	Chloramines	Sulfate	Conductivity	Organic_carbon	Trihalomethanes	Turbidity	Potability
count	2785.000000	3276.000000	3276.000000	3276.000000	2495.000000	3276.000000	3276.000000	3114.000000	3276.000000	3276.000000
mean	7.080795	196.369496	22014.092526	7.122277	333.775777	426.205111	14.284970	66.396293	3.966786	0.390110
std	1.594320	32.879761	8768.570828	1.583085	41.416840	80.824064	3.308162	16.175008	0.780382	0.487849
min	0.000000	47.432000	320.942611	0.352000	129.000000	181.483754	2.200000	0.738000	1.450000	0.000000
25%	6.093092	176.850538	15666.690297	6.127421	307.699498	365.734414	12.065801	55.844536	3.439711	0.000000
50%	7.036752	196.967627	20927.833607	7.130299	333.073546	421.884968	14.218338	66.622485	3.955028	0.000000
75%	8.062066	216.667456	27332.762127	8.114887	359.950170	481.792304	16.557652	77.337473	4.500320	1.000000
max	14.000000	323.124000	61227.196008	13.127000	481.030642	753.342620	28.300000	124.000000	6.739000	1.000000

dataframed.dtypes

ph                 float64
Hardness           float64
Solids             float64
Chloramines        float64
Sulfate            float64
Conductivity       float64
Organic_carbon     float64
Trihalomethanes    float64
Turbidity          float64
Potability           int64
dtype: object

dataframed.isnull().sum()

ph                 491
Hardness             0
Solids               0
Chloramines          0
Sulfate            781
Conductivity         0
Organic_carbon       0
Trihalomethanes    162
Turbidity            0
Potability           0
dtype: int64

Columns with missing values :

pH value: PH is an important parameter in evaluating the acid–base balance of water. It is also the indicator of acidic or alkaline condition of water status. WHO has recommended maximum permissible limit of pH from 6.5 to 8.5. The current investigation ranges were 6.52–6.83 which are in the range of WHO standards.
Sulfate: Sulfates are naturally occurring substances that are found in minerals, soil, and rocks. They are present in ambient air, groundwater, plants, and food. The principal commercial use of sulfate is in the chemical industry. Sulfate concentration in seawater is about 2,700 milligrams per liter (mg/L). It ranges from 3 to 30 mg/L in most freshwater supplies, although much higher concentrations (1000 mg/L) are found in some geographic locations.
Trihalomethanes: THMs are chemicals which may be found in water treated with chlorine. The concentration of THMs in drinking water varies according to the level of organic material in the water, the amount of chlorine required to treat the water, and the temperature of the water that is being treated. THM levels up to 80 ppm is considered safe in drinking water.

import missingno as msno
msno.matrix(dataframed)
plt.show()

Missing Values : PH Value

Scientists measure the hardness of water using a pH scale, which measures the hydrogen-ion concentration in the liquid. Water with a low pH is more acidic, while water with a higher pH is harder or more alkaline, meaning it is able to neutralize acids.

The pH scale measures substances on a scale from 1 to 14, with 7 being neutral.

Reference : Howell, E. (2013, March 26). What makes water hard?. LiveScience. https://www.livescience.com/34462-water-hard-ph.html

print('Conditonal Statements to fill in the Missing Values of PH Value Column')

print("\n")
print('if Potability = 0 and Hardness <= 150')
condition_1_mean_ph = dataframed[(dataframed['Potability'] == 0) & (dataframed['Hardness'] <= 150)][['ph']].mean()
print("PH VALUE : {:.4f}".format(float(condition_1_mean_ph)))

print("\n")
print('if Potability = 0 and Hardness > 150')
condition_2_mean_ph = dataframed[(dataframed['Potability'] == 0) & (dataframed['Hardness'] > 150)][['ph']].mean()
print("PH VALUE : {:.4f}".format(float(condition_2_mean_ph)))


print("\n")
print('if Potability = 1 and Hardness <= 150')
condition_3_mean_ph = dataframed[(dataframed['Potability'] == 0) & (dataframed['Hardness'] <= 150)][['ph']].mean()
print("PH VALUE : {:.4f}".format(float(condition_3_mean_ph)))

print("\n")
print('if Potability = 1 and Hardness > 150')
condition_4_mean_ph = dataframed[(dataframed['Potability'] == 0) & (dataframed['Hardness'] > 150)][['ph']].mean()
print("PH VALUE : {:.4f}".format(float(condition_4_mean_ph)))

Conditonal Statements to fill in the Missing Values of PH Value Column


if Potability = 0 and Hardness <= 150
PH VALUE : 6.7220


if Potability = 0 and Hardness > 150
PH VALUE : 7.1125


if Potability = 1 and Hardness <= 150
PH VALUE : 6.7220


if Potability = 1 and Hardness > 150
PH VALUE : 7.1125

for x in range(0, len(dataframed)) :
    if (pd.isnull(dataframed['ph'][x]) == True) :
        if ((dataframed['Potability'][x] == 0) & (dataframed['Hardness'][x] <= 150)) : dataframed['ph'][x] = condition_1_mean_ph
        elif ((dataframed['Potability'][x] == 0) & (dataframed['Hardness'][x] > 150)) : dataframed['ph'][x] = condition_2_mean_ph
        elif ((dataframed['Potability'][x] == 1) & (dataframed['Hardness'][x] <= 150)) : dataframed['ph'][x] = condition_3_mean_ph
        elif ((dataframed['Potability'][x] == 1) & (dataframed['Hardness'][x] > 150)) : dataframed['ph'][x] = condition_4_mean_ph

Missing Values : Sulfate

Seawater contains roughly 2,700 milligrammes per litre (mg/L) of sulphate. Most freshwater supplies have concentrations of 3 to 30 mg/L, while some geographic regions have significantly greater quantities (1000 mg/L).

Note : Sadly the Dataset only contains data of Sulfate ranging from 129 to 481 (mg/L) so we will just get mean of the rows that are potable and not potable which has a difference of 2 (mg/L)

print('Conditonal Statements to fill in the Missing Values of Sulfate Column')

print("\n")
print('if Potability = 0')
condition_1_mean_sulfate = dataframed[(dataframed['Potability'] == 0)][['Sulfate']].mean()
print("Sulfate : {:.4f}".format(float(condition_1_mean_sulfate)))

print("\n")
print('if Potability = 1')
condition_2_mean_sulfate = dataframed[(dataframed['Potability'] == 1)][['Sulfate']].mean()
print("Sulfate : {:.4f}".format(float(condition_2_mean_sulfate)))

Conditonal Statements to fill in the Missing Values of Sulfate Column


if Potability = 0
Sulfate : 334.5643


if Potability = 1
Sulfate : 332.5670

for x in range(0, len(dataframed)) :
    if (pd.isnull(dataframed['Sulfate'][x]) == True) :
        if (dataframed['Potability'][x] == 0) : dataframed['Sulfate'][x] = condition_1_mean_sulfate
        else : dataframed['Sulfate'][x] = condition_2_mean_sulfate

Missing Values : Trihalomethanes

THMs are chemicals which may be found in water treated with chlorine. The concentration of THMs in drinking water varies according to the level of organic material in the water, the amount of chlorine required to treat the water, and the temperature of the water that is being treated. THM levels up to 80 ppm is considered safe in drinking water.

dataframed['Trihalomethanes'].fillna(value = dataframed['Trihalomethanes'].mean() , inplace = True)

EDA

Types of Rows :

Overall Water Samples
Potable Water Samples
Not Potable Water Samples

Types of Rows : Overall Water Samples

The Range of Values among different 10 columns

print("The Range of Values among different 10 columns")

print("Ph : {:.4f}".format(min(dataframed['ph'])) + " - {:.4f}".format(max(dataframed['ph'])))
print('\n')
print("Hardness : {:.4f}".format(min(dataframed['Hardness'])) + " - {:.4f}".format(max(dataframed['Hardness'])))
print('\n')
print("Solids : {:.4f}".format(min(dataframed['Solids'])) + " - {:.4f}".format(max(dataframed['Solids'])))
print('\n')
print("Chloramines : {:.4f}".format(min(dataframed['Chloramines'])) + " - {:.4f}".format(max(dataframed['Chloramines'])))
print('\n')
print("Sulfate : {:.4f}".format(min(dataframed['Sulfate'])) + " - {:.4f}".format(max(dataframed['Sulfate'])))
print('\n')
print("Conductivity : {:.4f}".format(min(dataframed['Conductivity'])) + " - {:.4f}".format(max(dataframed['Conductivity'])))
print('\n')
print("Organic Carbon : {:.4f}".format(min(dataframed['Organic_carbon'])) + " - {:.4f}".format(max(dataframed['Organic_carbon'])))
print('\n')
print("Trihalomethanes : {:.4f}".format(min(dataframed['Trihalomethanes'])) + " - {:.4f}".format(max(dataframed['Trihalomethanes'])))
print('\n')
print("Turbidity : {:.4f}".format(min(dataframed['Turbidity'])) + " - {:.4f}".format(max(dataframed['Turbidity'])))
print('\n')
print("Potability : {:.4f}".format(min(dataframed['Potability'])) + " - {:.4f}".format(max(dataframed['Potability'])))

The Range of Values among different 10 columns
Ph : 0.0000 - 14.0000


Hardness : 47.4320 - 323.1240


Solids : 320.9426 - 61227.1960


Chloramines : 0.3520 - 13.1270


Sulfate : 129.0000 - 481.0306


Conductivity : 181.4838 - 753.3426


Organic Carbon : 2.2000 - 28.3000


Trihalomethanes : 0.7380 - 124.0000


Turbidity : 1.4500 - 6.7390


Potability : 0.0000 - 1.0000

Distribution of Values among different 10 columns

plt.figure(figsize=(16,14))
for i,col in enumerate(dataframed.columns):
    plt.subplot(4,3,i+1)
    sns.kdeplot(data=dataframed[col])
    plt.tight_layout()

Pearson correlation coefficient Heatmap

This kind of correlation expresses a value from 1 to -1 to represent the strength of relativity among two different variables.

sns.set(rc={'figure.figsize':(11.7,8.27)})
Heated = sns.heatmap(dataframed.corr("pearson"),vmin=-1, vmax=1,cmap='viridis',annot=True, square=True)
Heated.set(xlabel = "Pearson Correlation Coefficient Heatmap")

[Text(0.5, 56.016874999999985, 'Pearson Correlation Coefficient Heatmap')]

sns.pairplot(dataframed,hue='Potability')

<seaborn.axisgrid.PairGrid at 0x7aaf20f258d0>

Check for Outliers among the different 10 columns

plt.figure(figsize=(16,14))
for i,col in enumerate(dataframed.columns):
    plt.subplot(4,3,i+1)
    sns.boxplot(data=dataframed[col])
    
    plt.tight_layout()

Distribution of rows in terms of PH Level

count_plotted = sns.histplot(x = 'ph', data = dataframed, kde = True, color = 'dodgerblue')
count_plotted.set(xlabel = "PH Level")
count_plotted.set(xlim = (0.0000, 14.0000))

[(0.0, 14.0)]

Distribution of rows in terms of PH Level

Outliers : Ph Level

boxxer_plot = sns.boxplot(x = 'ph', data = dataframed, color = 'dodgerblue')
boxxer_plot.set(xlim = (0.0000, 14.0000))
boxxer_plot.set(xlabel = 'PH Level')

[Text(0.5, 0, 'PH Level')]

How Many Outliers in Ph Level Column?

numpy_array = np.array(dataframed['ph'])
percentile_1, percentile_2 = np.percentile(numpy_array, [25, 75])
bounds = percentile_2 - percentile_1
lower_bound = percentile_1 - 1.5 * bounds
upper_bound = percentile_2 + 1.5 * bounds
outliers = numpy_array[(numpy_array < lower_bound) | (numpy_array > upper_bound)]
print("The Outliers are values less than {:.4f}".format(lower_bound), "or greater than {:.4f}".format(upper_bound))
print("There are" ,len(outliers), "outliers.")

The Outliers are values less than 3.8891 or greater than 10.2586
There are 142 outliers.

Distribution of rows in terms of Hardness

count_plotted = sns.histplot(x = 'Hardness', data = dataframed, kde = True, color = 'dodgerblue')
count_plotted.set(xlabel = "Hardness")
count_plotted.set(xlim = (47.4320, 323.1240))

[(47.432, 323.124)]

Distribution of rows in terms of Hardness

Outliers : Hardness

boxxer_plot = sns.boxplot(x = 'Hardness', data = dataframed, color = 'dodgerblue')
boxxer_plot.set(xlim = (47.4320, 323.1240))

[(47.432, 323.124)]

How Many Outliers in Hardness Column?

numpy_array = np.array(dataframed['Hardness'])
percentile_1, percentile_2 = np.percentile(numpy_array, [25, 75])
bounds = percentile_2 - percentile_1
lower_bound = percentile_1 - 1.5 * bounds
upper_bound = percentile_2 + 1.5 * bounds
outliers = numpy_array[(numpy_array < lower_bound) | (numpy_array > upper_bound)]
print("The Outliers are values less than {:.4f}".format(lower_bound), "or greater than {:.4f}".format(upper_bound))
print("There are" ,len(outliers), "outliers.")

The Outliers are values less than 117.1252 or greater than 276.3928
There are 83 outliers.

Column : Solids (Total dissolved solids – TDS)

Distribution of rows in terms of Solids

count_plotted = sns.histplot(x = 'Solids', data = dataframed, kde = True, color = 'dodgerblue')
count_plotted.set(xlabel = "Solids")
count_plotted.set(xlim = (320.9426, 61227.1960))

[(320.9426, 61227.196)]

Outliers : Solids

boxxer_plot = sns.boxplot(x = 'Solids', data = dataframed, color = 'dodgerblue')
boxxer_plot.set(xlim = (320.9426, 61227.1960))

[(320.9426, 61227.196)]

How Many Outliers in Solids Column?

numpy_array = np.array(dataframed['Solids'])
percentile_1, percentile_2 = np.percentile(numpy_array, [25, 75])
bounds = percentile_2 - percentile_1
lower_bound = percentile_1 - 1.5 * bounds
upper_bound = percentile_2 + 1.5 * bounds
outliers = numpy_array[(numpy_array < lower_bound) | (numpy_array > upper_bound)]
print("The Outliers are greater than {:.4f}".format(upper_bound))
print("There are" ,len(outliers), "outliers.")

The Outliers are greater than 44831.8699
There are 47 outliers

Distribution of rows in terms of Chloramines

count_plotted = sns.histplot(x = 'Chloramines', data = dataframed, kde = True, color = 'dodgerblue')
count_plotted.set(xlabel = "Chloramines")
count_plotted.set(xlim = (0.3520, 13.1270))

[(0.352, 13.127)]

Distribution of rows in terms of Chloramines

How Many Outliers in Chloramines Column?

numpy_array = np.array(dataframed['Chloramines'])
percentile_1, percentile_2 = np.percentile(numpy_array, [25, 75])
bounds = percentile_2 - percentile_1
lower_bound = percentile_1 - 1.5 * bounds
upper_bound = percentile_2 + 1.5 * bounds
outliers = numpy_array[(numpy_array < lower_bound) | (numpy_array > upper_bound)]
print("The Outliers are values less than {:.4f}".format(lower_bound), "or greater than {:.4f}".format(upper_bound))
print("There are" ,len(outliers), "outliers.")

The Outliers are values less than 3.1462 or greater than 11.0961
There are 61 outliers.

Distribution of rows in terms of Sulfate

count_plotted = sns.histplot(x = 'Sulfate', data = dataframed, kde = True, color = 'dodgerblue')
count_plotted.set(xlabel = "Sulfate")
count_plotted.set(xlim = (129.0000, 481.0306))

[(129.0, 481.0306)]

Distribution of rows in terms of Sulfate

Defining X

Should contain the following :

PH VALUE
HARDNESS
SOLIDS
CHLORAMINES
SULFATE
CONDUCTIVITY
ORGANIC_CARBON
TRIHALOMETHANES
TURBIDITY

X = dataframed.iloc[:, :-1]
X.head()

	ph	Hardness	Solids	Chloramines	Sulfate	Conductivity	Organic_carbon	Trihalomethanes	Turbidity
0	7.112512	204.890455	20791.318981	7.300212	368.516441	564.308654	10.379783	86.990970	2.963135
1	3.716080	129.422921	18630.057858	6.635246	334.564290	592.885359	15.180013	56.329076	4.500656
2	8.099124	224.236259	19909.541732	9.275884	334.564290	418.606213	16.868637	66.420093	3.055934
3	8.316766	214.373394	22018.417441	8.059332	356.886136	363.266516	18.436524	100.341674	4.628771
4	9.092223	181.101509	17978.986339	6.546600	310.135738	398.410813	11.558279	31.997993	4.075075

Defining y

Should contain the following :

POTABILITY

y = dataframed.iloc[:, -1:]
y.head()

Potability
0
0
0
0
0

Training and Testing

Supervised AL used :

K-Nearest Neighbors
Decision Tree
Random Forest
Logistic Regression
Support Vector Machine
Gaussian Naive Bayes

from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score , classification_report, ConfusionMatrixDisplay,precision_score,recall_score, f1_score,roc_auc_score,roc_curve

Split The Data for Training and Testing Purposes

Test Sizes

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

Shape of Data (20%)

print('Shape of Data (20%)')
print("X_train shape : ", X_train.shape)
print("y_train shape : ", y_train.shape)
print("X_test shape : ", X_test.shape)
print("y_test shape : ", y_test.shape)

Shape of Data (20%)
X_train shape :  (2620, 9)
y_train shape :  (2620, 1)
X_test shape :  (656, 9)
y_test shape :  (656, 1)

AL : K-Nearest Neighbors (20%)

from sklearn.neighbors import KNeighborsClassifier

knn = KNeighborsClassifier(n_neighbors = 2)
knn.fit(X_train, y_train.values.ravel())
training_prediction = knn.predict(X_train)
testing_prediction = knn.predict(X_test)

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : K-Nearest Neighbors (20%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : K-Nearest Neighbors (20%)


Training Model Performance Check
Accuracy Score : 0.7618
F1 Score : 0.7299
Precision Score : 0.8291
Recall Score : 0.7618


Testing Model Performance Check
Accuracy Score : 0.5899
F1 Score : 0.5284
Precision Score : 0.5284
Recall Score : 0.5899

'''from sklearn.model_selection import GridSearchCV

leaf_size = list(range(1,50))
n_neighbors = list(range(1,30))
p=[1,2]
#Convert to dictionary
hyperparameters = dict(leaf_size=leaf_size, n_neighbors=n_neighbors, p=p)
#Create new KNN object
knn_2 = KNeighborsClassifier()
#Use GridSearch
clf = GridSearchCV(knn_2, hyperparameters, cv=10)
#Fit the model
best_model = clf.fit(X_train, y_train)
#Print The value of best Hyperparameters
print('Best leaf_size:', best_model.best_estimator_.get_params()['leaf_size'])
print('Best p:', best_model.best_estimator_.get_params()['p'])
print('Best n_neighbors:', best_model.best_estimator_.get_params()['n_neighbors'])
'''

"from sklearn.model_selection import GridSearchCV\n\nleaf_size = list(range(1,50))\nn_neighbors = list(range(1,30))\np=[1,2]\n#Convert to dictionary\nhyperparameters = dict(leaf_size=leaf_size, n_neighbors=n_neighbors, p=p)\n#Create new KNN object\nknn_2 = KNeighborsClassifier()\n#Use GridSearch\nclf = GridSearchCV(knn_2, hyperparameters, cv=10)\n#Fit the model\nbest_model = clf.fit(X_train, y_train)\n#Print The value of best Hyperparameters\nprint('Best leaf_size:', best_model.best_estimator_.get_params()['leaf_size'])\nprint('Best p:', best_model.best_estimator_.get_params()['p'])\nprint('Best n_neighbors:', best_model.best_estimator_.get_params()['n_neighbors'])\n"

AL : Decision Tree (20%)

from sklearn.tree import DecisionTreeClassifier
from sklearn import tree

Decision_Tree = DecisionTreeClassifier(max_depth = 5)
Decision_Tree.fit(X_train, y_train.values.ravel())
training_prediction = Decision_Tree.predict(X_train)
testing_prediction = Decision_Tree.predict(X_test)

# Visualize Decision Tree
plt.figure(figsize = (25,20))
tree.plot_tree(Decision_Tree,
               feature_names =  dataframed.columns.tolist()[:-1],
               class_names = ["0", "1"],
               filled = True,
               precision = 5)
plt.show()

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : Decision Tree (20%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score {:.4f}'.format(training_accuracy))
print('F1 Score {:.4f}'.format(training_f1))
print('Precision Score {:.4f}'.format(training_precision))
print('Recall Score {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score {:.4f}'.format(testing_accuracy))
print('F1 Score {:.4f}'.format(testing_f1))
print('Precision Score {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Decision Tree (20%)


Training Model Performance Check
Accuracy Score 0.7485
F1 Score 0.7289
Precision Score 0.7650
Recall Score 0.7485


Testing Model Performance Check
Accuracy Score 0.7210
F1 Score 0.7005
Precision Score 0.7190
Recall Score : 0.7210

AL : Random Forest (20%)

from sklearn.ensemble import RandomForestClassifier

Random_Forest = RandomForestClassifier()
Random_Forest.fit(X_train, y_train.values.ravel())
training_prediction = Random_Forest.predict(X_train)
testing_prediction = Random_Forest.predict(X_test) 

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : Random Forest (20%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Random Forest (20%)


Training Model Performance Check
Accuracy Score : 1.0000
F1 Score : 1.0000
Precision Score : 1.0000
Recall Score : 1.0000


Testing Model Performance Check
Accuracy Score : 0.7591
F1 Score : 0.7488
Precision Score : 0.7570
Recall Score : 0.7591

AL : Logistic Regression (20%)

from sklearn.linear_model import LogisticRegression
Logistic_Regression = LogisticRegression()
Logistic_Regression.fit(X_train, y_train.values.ravel())
training_prediction = Logistic_Regression.predict(X_train)
testing_prediction = Logistic_Regression.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Logistic Regression (20%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Logistic Regression (20%)


Training Model Performance Check
Accuracy Score : 0.6061
F1 Score : 0.4846
Precision Score : 0.3944
Recall Score : 0.6280


Testing Model Performance Check
Accuracy Score : 0.6280
F1 Score : 0.7488
Precision Score : 0.7570
Recall Score : 0.7591

AL : Support Vector Machine (20%)

from sklearn import svm
SVM = svm.SVC()
SVM.fit(X_train, y_train.values.ravel())
training_prediction = SVM.predict(X_train)
testing_prediction = SVM.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Support Vector Machine (20%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Support Vector Machine (20%)


Training Model Performance Check
Accuracy Score : 0.6053
F1 Score : 0.4846
Precision Score : 0.3944
Recall Score : 0.6280


Testing Model Performance Check
Accuracy Score : 0.6280
F1 Score : 0.7488
Precision Score : 0.7570
Recall Score : 0.7591

AL : Gaussian Naive Bayes (20%)

from sklearn.naive_bayes import GaussianNB
NB = GaussianNB()
NB.fit(X_train, y_train.values.ravel())
training_prediction = NB.predict(X_train)
testing_prediction = NB.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Gaussian Naive Bayes (20%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Gaussian Naive Bayes (20%)


Training Model Performance Check
Accuracy Score : 0.6302
F1 Score : 0.5825
Precision Score : 0.5981
Recall Score : 0.6296


Testing Model Performance Check
Accuracy Score : 0.6296
F1 Score : 0.7488
Precision Score : 0.7570
Recall Score : 0.7591

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 42)

Shape of Data (30%)
X_train shape :  (2293, 9)
y_train shape :  (2293, 1)
X_test shape :  (983, 9)
y_test shape :  (983, 1)

AL : K-Nearest Neighbors (30%)

from sklearn.neighbors import KNeighborsClassifier

knn = KNeighborsClassifier(n_neighbors = 2)
knn.fit(X_train, y_train.values.ravel())
training_prediction = knn.predict(X_train)
testing_prediction = knn.predict(X_test)

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : K-Nearest Neighbors (30%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : K-Nearest Neighbors (30%)


Training Model Performance Check
Accuracy Score : 0.7505
F1 Score : 0.7150
Precision Score : 0.8236
Recall Score : 0.7505


Testing Model Performance Check
Accuracy Score : 0.6002
F1 Score : 0.5405
Precision Score : 0.5464
Recall Score : 0.6002

AL : Decision Tree (30%)

from sklearn.tree import DecisionTreeClassifier
Decision_Tree = DecisionTreeClassifier(max_depth = 5)
Decision_Tree.fit(X_train, y_train.values.ravel())
training_prediction = Decision_Tree.predict(X_train)
testing_prediction = Decision_Tree.predict(X_test)

# Visualize Decision Tree
plt.figure(figsize = (25,20))
tree.plot_tree(Decision_Tree,
               feature_names =  dataframed.columns.tolist()[:-1],
               class_names = ["0", "1"],
               filled = True,
               precision = 5)
plt.show()

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : Decision Tree (30%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score {:.4f}'.format(training_accuracy))
print('F1 Score {:.4f}'.format(training_f1))
print('Precision Score {:.4f}'.format(training_precision))
print('Recall Score {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score {:.4f}'.format(testing_accuracy))
print('F1 Score {:.4f}'.format(testing_f1))
print('Precision Score {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Decision Tree (30%)


Training Model Performance Check
Accuracy Score 0.7719
F1 Score 0.7660
Precision Score 0.7717
Recall Score 0.7719


Testing Model Performance Check
Accuracy Score 0.7406
F1 Score 0.7330
Precision Score 0.7352
Recall Score : 0.7406

AL : Random Forest (30%)

from sklearn.ensemble import RandomForestClassifier

Random_Forest = RandomForestClassifier()
Random_Forest.fit(X_train, y_train.values.ravel())
training_prediction = Random_Forest.predict(X_train)
testing_prediction = Random_Forest.predict(X_test) 

#Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

#Testing Metrics
testing_accuracy = accuracy_score(y_test, testing_prediction)
testing_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
testing_precision = precision_score(y_test, testing_prediction, average = 'weighted')
testing_recall = recall_score(y_test, testing_prediction, average = 'weighted')

print('AL : Random Forest (30%)')
print('\n')
print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Random Forest (30%)


Training Model Performance Check
Accuracy Score : 1.0000
F1 Score : 1.0000
Precision Score : 1.0000
Recall Score : 1.0000


Testing Model Performance Check
Accuracy Score : 0.7762
F1 Score : 0.7671
Precision Score : 0.7757
Recall Score : 0.7762

AL : Logistic Regression (30%)

from sklearn.linear_model import LogisticRegression
Logistic_Regression = LogisticRegression()
Logistic_Regression.fit(X_train, y_train.values.ravel())
training_prediction = Logistic_Regression.predict(X_train)
testing_prediction = Logistic_Regression.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Logistic Regression (30%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Logistic Regression (30%)


Training Model Performance Check
Accuracy Score : 0.6027
F1 Score : 0.4841
Precision Score : 0.3940
Recall Score : 0.6277


Testing Model Performance Check
Accuracy Score : 0.6277
F1 Score : 0.7671
Precision Score : 0.7757
Recall Score : 0.7762

AL : Support Vector Machine (30%)

from sklearn import svm
SVM = svm.SVC()
SVM.fit(X_train, y_train.values.ravel())
training_prediction = SVM.predict(X_train)
testing_prediction = SVM.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Support Vector Machine (30%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Support Vector Machine (30%)


Training Model Performance Check
Accuracy Score : 0.6023
F1 Score : 0.4841
Precision Score : 0.3940
Recall Score : 0.6277


Testing Model Performance Check
Accuracy Score : 0.6277
F1 Score : 0.7671
Precision Score : 0.7757
Recall Score : 0.7762

AL : Gaussian Naive Bayes (30%)

from sklearn.naive_bayes import GaussianNB
NB = GaussianNB()
NB.fit(X_train, y_train.values.ravel())
training_prediction = NB.predict(X_train)
testing_prediction = NB.predict(X_test)

# Training Metrics
training_accuracy = accuracy_score(y_train, training_prediction)
training_f1 = f1_score(y_train, training_prediction, average = 'weighted')
training_precision = precision_score(y_train, training_prediction, average = 'weighted')
training_recall = recall_score(y_train, training_prediction, average = 'weighted')

# Testing Metrics

testing_accuracy = accuracy_score(y_test, testing_prediction)
training_f1 = f1_score(y_test, testing_prediction, average = 'weighted')
training_precision = precision_score(y_test, testing_prediction, average = 'weighted')
training_recall = recall_score(y_test, testing_prediction, average = 'weighted')


print('AL : Gaussian Naive Bayes (30%)')
print('\n')

print('Training Model Performance Check')
print('Accuracy Score : {:.4f}'.format(training_accuracy))
print('F1 Score : {:.4f}'.format(training_f1))
print('Precision Score : {:.4f}'.format(training_precision))
print('Recall Score : {:.4f}'.format(training_recall))

print('\n')
print('Testing Model Performance Check')
print('Accuracy Score : {:.4f}'.format(testing_accuracy))
print('F1 Score : {:.4f}'.format(testing_f1))
print('Precision Score : {:.4f}'.format(testing_precision))
print('Recall Score : {:.4f}'.format(testing_recall))

AL : Gaussian Naive Bayes (30%)


Training Model Performance Check
Accuracy Score : 0.6263
F1 Score : 0.5812
Precision Score : 0.6023
Recall Score : 0.6328


Testing Model Performance Check
Accuracy Score : 0.6328
F1 Score : 0.7671
Precision Score : 0.7757
Recall Score : 0.7762

Important Notice for college students

If you’re a college student and have skills in programming languages, Want to earn through blogging? Mail us at geekycomail@gmail.com

For more Programming related blogs Visit Us Geekycodes. Follow us on Instagram

ML-Based Water Portability Prediction

CONTEXT

The Dataset consists of the following :

METADATA

Columns with missing values :

Missing Values : PH Value

Missing Values : Sulfate

Missing Values : Trihalomethanes

EDA

Types of Rows : Overall Water Samples

The Range of Values among different 10 columns

Pearson correlation coefficient Heatmap

Outliers : Ph Level

Column : Solids (Total dissolved solids – TDS)

Defining X

Defining y

Training and Testing

Split The Data for Training and Testing Purposes

AL : K-Nearest Neighbors (20%)

AL : Decision Tree (20%)

AL : Random Forest (20%)

AL : Logistic Regression (20%)

AL : Support Vector Machine (20%)

AL : Gaussian Naive Bayes (20%)

Important Notice for college students

Like this:

Related

Leave a ReplyCancel reply

CONTEXT

The Dataset consists of the following :

METADATA

Columns with missing values :

Missing Values : PH Value

Missing Values : Sulfate

Missing Values : Trihalomethanes

EDA

Types of Rows : Overall Water Samples

The Range of Values among different 10 columns

Pearson correlation coefficient Heatmap

Outliers : Ph Level

Column : Solids (Total dissolved solids – TDS)

Defining X

Defining y

Training and Testing

Split The Data for Training and Testing Purposes

AL : K-Nearest Neighbors (20%)

AL : Decision Tree (20%)

AL : Random Forest (20%)

AL : Logistic Regression (20%)

AL : Support Vector Machine (20%)

AL : Gaussian Naive Bayes (20%)

Important Notice for college students

Share this:

Like this:

Related

Leave a ReplyCancel reply