4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Import libraries
In [1]: import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import seaborn as sns
import itertools
import plotly.express as px
import plotly.graph_objects as go
import plotly.io as pio
from plotly.subplots import make_subplots
import warnings
warnings.filterwarnings('ignore')
# NN models
import keras
from keras.models import Sequential
from keras.layers import Dense, Dropout
from keras import optimizers
from keras.wrappers.scikit_learn import KerasClassifier
from keras.callbacks import EarlyStopping, ModelCheckpoint
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename))

/kaggle/input/milkquality/milknew.csv

Loading the dataset

In [2]: df=pd.read_csv('/kaggle/input/milkquality/milknew.csv')
df

Out[2]: pH Temprature Taste Odor Fat Turbidity Colour Grade

0 6.6 35 1 0 1 0 254 high

1 6.6 36 0 1 0 1 253 high

2 8.5 70 1 1 1 1 246 low

3 9.5 34 1 1 0 1 255 low

4 6.6 37 0 0 0 0 255 medium

... ... ... ... ... ... ... ... ...

1054 6.7 45 1 1 0 0 247 medium

1055 6.7 38 1 0 1 0 255 high

1056 3.0 40 1 1 1 1 255 low

1057 6.8 43 1 0 1 0 250 high

1058 8.6 55 0 1 1 1 255 low

1059 rows × 8 columns

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 1/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Exploratory Data Analysis

In [3]: df.shape

Out[3]: (1059, 8)

In [4]: df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1059 entries, 0 to 1058
Data columns (total 8 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 pH 1059 non-null float64
1 Temprature 1059 non-null int64
2 Taste 1059 non-null int64
3 Odor 1059 non-null int64
4 Fat 1059 non-null int64
5 Turbidity 1059 non-null int64
6 Colour 1059 non-null int64
7 Grade 1059 non-null object
dtypes: float64(1), int64(6), object(1)
memory usage: 66.3+ KB

In [5]: df.isna().sum()

Out[5]: pH 0
Temprature 0
Taste 0
Odor 0
Fat 0
Turbidity 0
Colour 0
Grade 0
dtype: int64

In [6]: df.duplicated().sum()

Out[6]: 976

In [7]: df.nunique()

Out[7]: pH 16
Temprature 17
Taste 2
Odor 2
Fat 2
Turbidity 2
Colour 9
Grade 3
dtype: int64

In [8]: df['Grade'].value_counts()

Out[8]: low 429

medium 374
high 256
Name: Grade, dtype: int64

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 2/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [9]: #df['Grade'] =df['Grade'].map({'low': 0, 'medium': 1,'high':2})

In [10]: df.describe().T

Out[10]: count mean std min 25% 50% 75% max

pH 1059.0 6.630123 1.399679 3.0 6.5 6.7 6.8 9.5

Temprature 1059.0 44.226629 10.098364 34.0 38.0 41.0 45.0 90.0

Taste 1059.0 0.546742 0.498046 0.0 0.0 1.0 1.0 1.0

Odor 1059.0 0.432483 0.495655 0.0 0.0 0.0 1.0 1.0

Fat 1059.0 0.671388 0.469930 0.0 0.0 1.0 1.0 1.0

Turbidity 1059.0 0.491029 0.500156 0.0 0.0 0.0 1.0 1.0

Colour 1059.0 251.840415 4.307424 240.0 250.0 255.0 255.0 255.0

Data Visualization
In [11]: for i in df.columns:
plt.figure(figsize=(13,7))
sns.histplot(data = df[i], kde=True, multiple='stack')
plt.xticks(rotation=90)
plt.show()

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 3/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 4/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 5/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [12]: plt.style.use("seaborn")
df.hist(figsize=(25,20), bins=15)

Out[12]: array([[<AxesSubplot:title={'center':'pH'}>,
<AxesSubplot:title={'center':'Temprature'}>,
<AxesSubplot:title={'center':'Taste'}>],
[<AxesSubplot:title={'center':'Odor'}>,
<AxesSubplot:title={'center':'Fat '}>,
<AxesSubplot:title={'center':'Turbidity'}>],
[<AxesSubplot:title={'center':'Colour'}>, <AxesSubplot:>,
<AxesSubplot:>]], dtype=object)

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 6/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [13]: plt.figure(figsize=(15,6))
sns.barplot(data = df, color = 'lightblue')
plt.xticks(rotation=90, fontsize=10)
plt.show()

In [14]: plt.figure(figsize=(15,6))
sns.boxenplot(data = df, color = 'lightblue')
plt.xticks(rotation=90, fontsize=10)
plt.show()

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 7/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [15]: sns.kdeplot(x=df["Odor"], y=df["Temprature"], hue =df["Grade"], palette="crest");

plt.show()

In [16]: sns.pairplot(df,hue='Grade')

Out[16]: <seaborn.axisgrid.PairGrid at 0x71c837028550>

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 8/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [17]: plt.figure(figsize=(35, 25))

corr = df.corr()
mask = np.triu(np.ones_like(df.corr(), dtype=np.bool))
heatmap = sns.heatmap(corr, mask = mask, vmin=-1, vmax=1, annot=True, cmap = 'virid
heatmap.set_title('Correlation Heatmap', fontdict={'fontsize':20}, pad=12)

Out[17]: Text(0.5, 1.0, 'Correlation Heatmap')

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 9/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [18]: df['Grade'] =df['Grade'].map({'low': 0, 'medium': 1,'high':2})

Data Preparation
In [19]: x=df.drop('Grade',axis=1)
y=df['Grade']

Model Building
In [20]: from sklearn.model_selection import train_test_split
x_train,x_test,y_train,y_test = train_test_split(x,y,test_size = 0.30, random_state

In [21]: from sklearn.preprocessing import StandardScaler

ss = StandardScaler()

x_train_std = ss.fit_transform(x_train)
x_test_std = ss.transform(x_test)

SVM
In [22]: from sklearn import metrics
from sklearn.model_selection import GridSearchCV, RandomizedSearchCV
from sklearn.svm import SVC

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 10/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

svc=SVC()
svc.fit(x_train,y_train)

Out[22]: SVC()

In [23]: print("Training Accuracy :",svc.score(x_train,y_train))

print("Testing Accuracy :",svc.score(x_test,y_test))

Training Accuracy : 0.5479082321187584

Testing Accuracy : 0.5345911949685535

In [24]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_svc = svc.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_svc)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for SVM", fontsize=10, y=1.03)

Out[24]: Text(0.5, 1.03, 'Confusion Matrix for SVM')

In [25]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_svc))

precision recall f1-score support

0 1.00 0.47 0.64 116

1 0.44 1.00 0.61 116
2 0.00 0.00 0.00 86

accuracy 0.53 318

macro avg 0.48 0.49 0.42 318
weighted avg 0.53 0.53 0.45 318

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 11/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Decision Tree
In [26]: from sklearn.tree import DecisionTreeClassifier
dtc=DecisionTreeClassifier(criterion='entropy',random_state=42)
dtc.fit(x_train,y_train)

Out[26]: DecisionTreeClassifier(criterion='entropy', random_state=42)

In [27]: print("Training Accuracy :",dtc.score(x_train,y_train))

print("Testing Accuracy :",dtc.score(x_test,y_test))

Training Accuracy : 1.0

Testing Accuracy : 0.9874213836477987

In [28]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_dtc = dtc.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_dtc)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for DecisionTreeClassifier", fontsize=10, y=1.03)

Out[28]: Text(0.5, 1.03, 'Confusion Matrix for DecisionTreeClassifier')

In [29]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_dtc))

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 12/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

precision recall f1-score support

0 1.00 0.97 0.99 116

1 0.98 1.00 0.99 116
2 0.98 0.99 0.98 86

accuracy 0.99 318

macro avg 0.99 0.99 0.99 318
weighted avg 0.99 0.99 0.99 318

Random Forest
In [30]: from sklearn.ensemble import RandomForestClassifier
rfc=RandomForestClassifier()
rfc.fit(x_train,y_train)

Out[30]: RandomForestClassifier()

In [31]: print("Training Accuracy :",rfc.score(x_train,y_train))

print("Testing Accuracy :",rfc.score(x_test,y_test))

Training Accuracy : 1.0

Testing Accuracy : 0.9874213836477987

In [32]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_rfc = rfc.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_rfc)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for RandomForestClassifier", fontsize=10, y=1.03)

Out[32]: Text(0.5, 1.03, 'Confusion Matrix for RandomForestClassifier')

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 13/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [33]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_rfc))

precision recall f1-score support

0 1.00 0.97 0.99 116

1 0.98 1.00 0.99 116
2 0.98 0.99 0.98 86

accuracy 0.99 318

macro avg 0.99 0.99 0.99 318
weighted avg 0.99 0.99 0.99 318

Gaussian Naive Bias

In [34]: from sklearn.naive_bayes import GaussianNB
nb = GaussianNB()
nb.fit(x_train, y_train)

Out[34]: GaussianNB()

In [35]: print("Training Accuracy :",nb.score(x_train,y_train))

print("Testing Accuracy :",nb.score(x_test,y_test))

Training Accuracy : 0.8421052631578947

Testing Accuracy : 0.8081761006289309

In [36]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_nb = nb.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_nb)
localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 14/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for GaussianNB", fontsize=10, y=1.03)

Out[36]: Text(0.5, 1.03, 'Confusion Matrix for GaussianNB')

In [37]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_nb))

precision recall f1-score support

0 0.99 0.93 0.96 116

1 0.97 0.55 0.70 116
2 0.59 0.99 0.74 86

accuracy 0.81 318

macro avg 0.85 0.82 0.80 318
weighted avg 0.88 0.81 0.81 318

Linear Regression
In [38]: from sklearn.linear_model import LinearRegression
lnr = LinearRegression()
lnr.fit(x_train, y_train)

Out[38]: LinearRegression()

In [39]: print("Training Accuracy :",lnr.score(x_train,y_train))

print("Testing Accuracy :",lnr.score(x_test,y_test))

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 15/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Training Accuracy : 0.2797660354192826

Testing Accuracy : 0.2543546562708897

Logistic Regression
In [40]: from sklearn.linear_model import LogisticRegression
lr=LogisticRegression()
lr.fit(x_train,y_train)

Out[40]: LogisticRegression()

In [41]: print("Training Accuracy :",lr.score(x_train,y_train))

print("Testing Accuracy :",lr.score(x_test,y_test))

Training Accuracy : 0.7462887989203779

Testing Accuracy : 0.7327044025157232

In [42]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_lr = lr.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_lr)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for Logistic Regression", fontsize=10, y=1.03)

Out[42]: Text(0.5, 1.03, 'Confusion Matrix for Logistic Regression')

In [43]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_lr))

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 16/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

precision recall f1-score support

0 0.70 0.81 0.75 116

1 0.78 0.78 0.78 116
2 0.72 0.57 0.64 86

accuracy 0.73 318

macro avg 0.73 0.72 0.72 318
weighted avg 0.73 0.73 0.73 318

XgBoost
In [44]: from xgboost import XGBClassifier
xgb = XGBClassifier()
xgb.fit(x_train, y_train)

Out[44]: XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,

colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,
early_stopping_rounds=None, enable_categorical=False,
eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',
importance_type=None, interaction_constraints='',
learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,
max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,
missing=nan, monotone_constraints='()', n_estimators=100,
n_jobs=0, num_parallel_tree=1, objective='multi:softprob',
predictor='auto', random_state=0, reg_alpha=0, ...)

In [45]: print("Training Accuracy :",xgb.score(x_train,y_train))

print("Testing Accuracy :",xgb.score(x_test,y_test))

Training Accuracy : 1.0

Testing Accuracy : 0.9874213836477987

In [46]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_xgb = xgb.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_xgb)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for XGBClassifierClassifier", fontsize=10, y=1.03)

Out[46]: Text(0.5, 1.03, 'Confusion Matrix for XGBClassifierClassifier')

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 17/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

In [47]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_xgb))

precision recall f1-score support

0 1.00 0.97 0.99 116

1 0.98 1.00 0.99 116
2 0.98 0.99 0.98 86

accuracy 0.99 318

macro avg 0.99 0.99 0.99 318
weighted avg 0.99 0.99 0.99 318

KNeighbors
In [48]: from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier()
knn.fit(x_train,y_train)

Out[48]: KNeighborsClassifier()

In [49]: print("Training Accuracy :",knn.score(x_train,y_train))

print("Testing Accuracy :",knn.score(x_test,y_test))

Training Accuracy : 0.9919028340080972

Testing Accuracy : 0.9874213836477987

In [50]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_knn = knn.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_knn)
localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 18/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for KNeighborsClassifier", fontsize=10, y=1.03)

Out[50]: Text(0.5, 1.03, 'Confusion Matrix for KNeighborsClassifier')

In [51]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_knn))

precision recall f1-score support

0 1.00 0.97 0.99 116

1 0.97 0.99 0.98 116
2 0.99 1.00 0.99 86

accuracy 0.99 318

macro avg 0.99 0.99 0.99 318
weighted avg 0.99 0.99 0.99 318

GradientBoosting
In [52]: from sklearn.ensemble import GradientBoostingClassifier
gb = GradientBoostingClassifier()
gb.fit(x_train, y_train)

Out[52]: GradientBoostingClassifier()

In [53]: print("Training Accuracy :",gb.score(x_train,y_train))

print("Testing Accuracy :",gb.score(x_test,y_test))

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 19/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Training Accuracy : 1.0

Testing Accuracy : 0.9905660377358491

In [54]: from sklearn.metrics import confusion_matrix, classification_report

y_pred_gb = gb.predict(x_test)
cf_matrix = confusion_matrix(y_test, y_pred_gb)
sns.heatmap(cf_matrix, annot=True)
plt.title("Confusion Matrix for GradientBoostingClassifier", fontsize=10, y=1.03)

Out[54]: Text(0.5, 1.03, 'Confusion Matrix for GradientBoostingClassifier')

In [55]: from sklearn import metrics

print(metrics.classification_report(y_test, y_pred_gb))

precision recall f1-score support

0 1.00 0.97 0.99 116

1 0.99 1.00 1.00 116
2 0.98 1.00 0.99 86

accuracy 0.99 318

macro avg 0.99 0.99 0.99 318
weighted avg 0.99 0.99 0.99 318

Artificial neural network

In [56]: # ANN Model Layers
from tensorflow.keras.layers import BatchNormalization
ann_model =Sequential()

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 20/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

ann_model.add(Dense(units = 32,activation = 'relu'))

ann_model.add(BatchNormalization())
ann_model.add(Dropout(0.5))

ann_model.add(Dense(units = 64,activation = 'relu'))

ann_model.add(BatchNormalization())
ann_model.add(Dropout(0.5))

ann_model.add(Dense(units = 1,activation = 'sigmoid'))

# Model Optimizer
ann_model.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['acc

In [57]: from tensorflow.keras.callbacks import EarlyStopping

early_stop = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=25)

In [58]: # Training the ANN

history = ann_model.fit(x_train, y_train, batch_size=16, epochs=100, validation_dat

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 21/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Epoch 1/100
47/47 [==============================] - 2s 8ms/step - loss: 0.9364 - accuracy: 0.
4521 - val_loss: 6.3271 - val_accuracy: 0.3648
Epoch 2/100
47/47 [==============================] - 0s 3ms/step - loss: 0.4443 - accuracy: 0.
5412 - val_loss: 1.0002 - val_accuracy: 0.4528
Epoch 3/100
47/47 [==============================] - 0s 3ms/step - loss: 0.0975 - accuracy: 0.
5776 - val_loss: 0.7874 - val_accuracy: 0.4686
Epoch 4/100
47/47 [==============================] - 0s 4ms/step - loss: 0.0673 - accuracy: 0.
5344 - val_loss: -0.0601 - val_accuracy: 0.5252
Epoch 5/100
47/47 [==============================] - 0s 3ms/step - loss: -0.0162 - accuracy:
0.5574 - val_loss: 1.2288 - val_accuracy: 0.4088
Epoch 6/100
47/47 [==============================] - 0s 4ms/step - loss: -0.1415 - accuracy:
0.5762 - val_loss: 4.8013 - val_accuracy: 0.3648
Epoch 7/100
47/47 [==============================] - 0s 5ms/step - loss: -0.3351 - accuracy:
0.5803 - val_loss: 1.4062 - val_accuracy: 0.3742
Epoch 8/100
47/47 [==============================] - 0s 3ms/step - loss: -0.4995 - accuracy:
0.5830 - val_loss: 1.0389 - val_accuracy: 0.4119
Epoch 9/100
47/47 [==============================] - 0s 3ms/step - loss: -0.6944 - accuracy:
0.5843 - val_loss: -0.2170 - val_accuracy: 0.5629
Epoch 10/100
47/47 [==============================] - 0s 3ms/step - loss: -0.8488 - accuracy:
0.5870 - val_loss: -0.0185 - val_accuracy: 0.5597
Epoch 11/100
47/47 [==============================] - 0s 3ms/step - loss: -0.8800 - accuracy:
0.5695 - val_loss: -0.8343 - val_accuracy: 0.4874
Epoch 12/100
47/47 [==============================] - 0s 4ms/step - loss: -1.2688 - accuracy:
0.5843 - val_loss: -0.7517 - val_accuracy: 0.5126
Epoch 13/100
47/47 [==============================] - 0s 4ms/step - loss: -1.3348 - accuracy:
0.5884 - val_loss: 4.9570 - val_accuracy: 0.2799
Epoch 14/100
47/47 [==============================] - 0s 3ms/step - loss: -1.7121 - accuracy:
0.5735 - val_loss: 7.0937 - val_accuracy: 0.3648
Epoch 15/100
47/47 [==============================] - 0s 4ms/step - loss: -2.0444 - accuracy:
0.5816 - val_loss: 6.9729 - val_accuracy: 0.3648
Epoch 16/100
47/47 [==============================] - 0s 4ms/step - loss: -2.3847 - accuracy:
0.5951 - val_loss: 12.9911 - val_accuracy: 0.3648
Epoch 17/100
47/47 [==============================] - 0s 4ms/step - loss: -2.7639 - accuracy:
0.5789 - val_loss: 2.9185 - val_accuracy: 0.4057
Epoch 18/100
47/47 [==============================] - 0s 4ms/step - loss: -2.8672 - accuracy:
0.5695 - val_loss: -1.0519 - val_accuracy: 0.6635
Epoch 19/100
47/47 [==============================] - 0s 4ms/step - loss: -3.5181 - accuracy:
0.5830 - val_loss: 0.0758 - val_accuracy: 0.4654
Epoch 20/100
47/47 [==============================] - 0s 4ms/step - loss: -3.7408 - accuracy:
0.5682 - val_loss: -1.5211 - val_accuracy: 0.6006
Epoch 21/100

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 22/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

47/47 [==============================] - 0s 3ms/step - loss: -4.0925 - accuracy:

0.5655 - val_loss: 0.6326 - val_accuracy: 0.4434
Epoch 22/100
47/47 [==============================] - 0s 3ms/step - loss: -5.1065 - accuracy:
0.5506 - val_loss: -1.1901 - val_accuracy: 0.5597
Epoch 23/100
47/47 [==============================] - 0s 4ms/step - loss: -5.2625 - accuracy:
0.5628 - val_loss: -2.6385 - val_accuracy: 0.6069
Epoch 24/100
47/47 [==============================] - 0s 3ms/step - loss: -5.7392 - accuracy:
0.5776 - val_loss: -3.2064 - val_accuracy: 0.5629
Epoch 25/100
47/47 [==============================] - 0s 4ms/step - loss: -5.7975 - accuracy:
0.5493 - val_loss: -6.1050 - val_accuracy: 0.5157
Epoch 26/100
47/47 [==============================] - 0s 3ms/step - loss: -6.4957 - accuracy:
0.5601 - val_loss: -6.2278 - val_accuracy: 0.4874
Epoch 27/100
47/47 [==============================] - 0s 4ms/step - loss: -7.2969 - accuracy:
0.5628 - val_loss: -6.4301 - val_accuracy: 0.5283
Epoch 28/100
47/47 [==============================] - 0s 4ms/step - loss: -7.9232 - accuracy:
0.5560 - val_loss: -5.6027 - val_accuracy: 0.4308
Epoch 29/100
47/47 [==============================] - 0s 4ms/step - loss: -8.0299 - accuracy:
0.5466 - val_loss: -4.8958 - val_accuracy: 0.3994
Epoch 30/100
47/47 [==============================] - 0s 4ms/step - loss: -7.6518 - accuracy:
0.5371 - val_loss: -5.8275 - val_accuracy: 0.4528
Epoch 31/100
47/47 [==============================] - 0s 4ms/step - loss: -8.2209 - accuracy:
0.5452 - val_loss: -5.5315 - val_accuracy: 0.3648
Epoch 32/100
47/47 [==============================] - 0s 3ms/step - loss: -9.4250 - accuracy:
0.5439 - val_loss: -9.9199 - val_accuracy: 0.4528
Epoch 33/100
47/47 [==============================] - 0s 4ms/step - loss: -9.1802 - accuracy:
0.5520 - val_loss: -10.9591 - val_accuracy: 0.4465
Epoch 34/100
47/47 [==============================] - 0s 4ms/step - loss: -11.6490 - accuracy:
0.5439 - val_loss: -15.8427 - val_accuracy: 0.4623
Epoch 35/100
47/47 [==============================] - 0s 4ms/step - loss: -11.1387 - accuracy:
0.5385 - val_loss: -9.5735 - val_accuracy: 0.6384
Epoch 36/100
47/47 [==============================] - 0s 4ms/step - loss: -13.5299 - accuracy:
0.5587 - val_loss: 10.5220 - val_accuracy: 0.3553
Epoch 37/100
47/47 [==============================] - 0s 3ms/step - loss: -15.1275 - accuracy:
0.5479 - val_loss: -18.1456 - val_accuracy: 0.4497
Epoch 38/100
47/47 [==============================] - 0s 3ms/step - loss: -14.3272 - accuracy:
0.5479 - val_loss: -20.7327 - val_accuracy: 0.5912
Epoch 39/100
47/47 [==============================] - 0s 4ms/step - loss: -14.6484 - accuracy:
0.5520 - val_loss: 0.2331 - val_accuracy: 0.3994
Epoch 40/100
47/47 [==============================] - 0s 4ms/step - loss: -15.2954 - accuracy:
0.5506 - val_loss: 9.7962 - val_accuracy: 0.3931
Epoch 41/100
47/47 [==============================] - 0s 4ms/step - loss: -17.3441 - accuracy:

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 23/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

0.5452 - val_loss: 4.0273 - val_accuracy: 0.4057

Epoch 42/100
47/47 [==============================] - 0s 3ms/step - loss: -17.4842 - accuracy:
0.5385 - val_loss: -12.3480 - val_accuracy: 0.5975
Epoch 43/100
47/47 [==============================] - 0s 4ms/step - loss: -17.4094 - accuracy:
0.5709 - val_loss: -9.5445 - val_accuracy: 0.5220
Epoch 44/100
47/47 [==============================] - 0s 3ms/step - loss: -20.2418 - accuracy:
0.5560 - val_loss: -34.2692 - val_accuracy: 0.4874
Epoch 45/100
47/47 [==============================] - 0s 4ms/step - loss: -20.9230 - accuracy:
0.5601 - val_loss: -22.2479 - val_accuracy: 0.5975
Epoch 46/100
47/47 [==============================] - 0s 3ms/step - loss: -22.7544 - accuracy:
0.5668 - val_loss: -29.6942 - val_accuracy: 0.5440
Epoch 47/100
47/47 [==============================] - 0s 3ms/step - loss: -24.4336 - accuracy:
0.5506 - val_loss: -33.8989 - val_accuracy: 0.4277
Epoch 48/100
47/47 [==============================] - 0s 3ms/step - loss: -22.6367 - accuracy:
0.5425 - val_loss: -35.0027 - val_accuracy: 0.3774
Epoch 49/100
47/47 [==============================] - 0s 3ms/step - loss: -25.2419 - accuracy:
0.5425 - val_loss: -23.7054 - val_accuracy: 0.5220
Epoch 50/100
47/47 [==============================] - 0s 4ms/step - loss: -29.5448 - accuracy:
0.5614 - val_loss: -36.6991 - val_accuracy: 0.4497
Epoch 51/100
47/47 [==============================] - 0s 3ms/step - loss: -28.5572 - accuracy:
0.5479 - val_loss: -23.5705 - val_accuracy: 0.3774
Epoch 52/100
47/47 [==============================] - 0s 3ms/step - loss: -29.2854 - accuracy:
0.5479 - val_loss: -36.7628 - val_accuracy: 0.6132
Epoch 53/100
47/47 [==============================] - 0s 3ms/step - loss: -30.0023 - accuracy:
0.5358 - val_loss: 37.9504 - val_accuracy: 0.3113
Epoch 54/100
47/47 [==============================] - 0s 4ms/step - loss: -31.3725 - accuracy:
0.5398 - val_loss: -16.2135 - val_accuracy: 0.5031
Epoch 55/100
47/47 [==============================] - 0s 3ms/step - loss: -33.4582 - accuracy:
0.5439 - val_loss: -29.0869 - val_accuracy: 0.6069
Epoch 56/100
47/47 [==============================] - 0s 3ms/step - loss: -36.7839 - accuracy:
0.5668 - val_loss: -58.7504 - val_accuracy: 0.4277
Epoch 57/100
47/47 [==============================] - 0s 4ms/step - loss: -32.0817 - accuracy:
0.5344 - val_loss: -46.9754 - val_accuracy: 0.6069
Epoch 58/100
47/47 [==============================] - 0s 3ms/step - loss: -36.9227 - accuracy:
0.5574 - val_loss: -39.8727 - val_accuracy: 0.4906
Epoch 59/100
47/47 [==============================] - 0s 3ms/step - loss: -41.1633 - accuracy:
0.5479 - val_loss: -43.5720 - val_accuracy: 0.4780
Epoch 60/100
47/47 [==============================] - 0s 4ms/step - loss: -38.3367 - accuracy:
0.5466 - val_loss: -54.9898 - val_accuracy: 0.4497
Epoch 61/100
47/47 [==============================] - 0s 3ms/step - loss: -43.9839 - accuracy:
0.5398 - val_loss: -11.2951 - val_accuracy: 0.5566

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 24/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Epoch 62/100
47/47 [==============================] - 0s 3ms/step - loss: -43.1137 - accuracy:
0.5385 - val_loss: 49.5336 - val_accuracy: 0.3648
Epoch 63/100
47/47 [==============================] - 0s 3ms/step - loss: -44.0907 - accuracy:
0.5412 - val_loss: -66.1592 - val_accuracy: 0.4874
Epoch 64/100
47/47 [==============================] - 0s 3ms/step - loss: -42.1352 - accuracy:
0.5439 - val_loss: -49.3381 - val_accuracy: 0.5535
Epoch 65/100
47/47 [==============================] - 0s 3ms/step - loss: -45.9768 - accuracy:
0.5547 - val_loss: -6.1029 - val_accuracy: 0.5377
Epoch 66/100
47/47 [==============================] - 0s 3ms/step - loss: -51.6089 - accuracy:
0.5601 - val_loss: -78.2757 - val_accuracy: 0.4937
Epoch 67/100
47/47 [==============================] - 0s 4ms/step - loss: -49.7343 - accuracy:
0.5506 - val_loss: -69.9387 - val_accuracy: 0.5472
Epoch 68/100
47/47 [==============================] - 0s 3ms/step - loss: -43.6494 - accuracy:
0.5439 - val_loss: -51.1160 - val_accuracy: 0.6069
Epoch 69/100
47/47 [==============================] - 0s 4ms/step - loss: -50.6234 - accuracy:
0.5250 - val_loss: -67.5859 - val_accuracy: 0.4245
Epoch 70/100
47/47 [==============================] - 0s 3ms/step - loss: -49.0533 - accuracy:
0.5371 - val_loss: -77.0359 - val_accuracy: 0.4151
Epoch 71/100
47/47 [==============================] - 0s 3ms/step - loss: -54.8261 - accuracy:
0.5425 - val_loss: -81.6451 - val_accuracy: 0.4119
Epoch 72/100
47/47 [==============================] - 0s 3ms/step - loss: -54.1311 - accuracy:
0.5439 - val_loss: -45.4728 - val_accuracy: 0.5409
Epoch 73/100
47/47 [==============================] - 0s 4ms/step - loss: -58.9882 - accuracy:
0.5560 - val_loss: 32.9783 - val_accuracy: 0.3302
Epoch 74/100
47/47 [==============================] - 0s 3ms/step - loss: -56.2245 - accuracy:
0.5547 - val_loss: -106.2492 - val_accuracy: 0.4717
Epoch 75/100
47/47 [==============================] - 0s 4ms/step - loss: -53.6323 - accuracy:
0.5479 - val_loss: -67.5443 - val_accuracy: 0.3648
Epoch 76/100
47/47 [==============================] - 0s 3ms/step - loss: -54.8622 - accuracy:
0.5398 - val_loss: -54.1232 - val_accuracy: 0.6164
Epoch 77/100
47/47 [==============================] - 0s 3ms/step - loss: -59.2817 - accuracy:
0.5277 - val_loss: -72.1177 - val_accuracy: 0.3648
Epoch 78/100
47/47 [==============================] - 0s 4ms/step - loss: -60.6974 - accuracy:
0.5290 - val_loss: -95.8444 - val_accuracy: 0.4780
Epoch 79/100
47/47 [==============================] - 0s 4ms/step - loss: -58.1961 - accuracy:
0.5290 - val_loss: -115.4964 - val_accuracy: 0.4811
Epoch 80/100
47/47 [==============================] - 0s 4ms/step - loss: -65.8021 - accuracy:
0.5425 - val_loss: -108.7465 - val_accuracy: 0.4025
Epoch 81/100
47/47 [==============================] - 0s 3ms/step - loss: -69.1130 - accuracy:
0.5398 - val_loss: -96.1030 - val_accuracy: 0.4025
Epoch 82/100

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 25/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

47/47 [==============================] - 0s 3ms/step - loss: -68.1573 - accuracy:

0.5398 - val_loss: -136.7848 - val_accuracy: 0.4686
Epoch 83/100
47/47 [==============================] - 0s 4ms/step - loss: -72.5733 - accuracy:
0.5520 - val_loss: 50.5330 - val_accuracy: 0.3994
Epoch 84/100
47/47 [==============================] - 0s 3ms/step - loss: -66.6355 - accuracy:
0.5385 - val_loss: -54.6375 - val_accuracy: 0.6541
Epoch 85/100
47/47 [==============================] - 0s 3ms/step - loss: -80.4892 - accuracy:
0.5385 - val_loss: -139.3182 - val_accuracy: 0.4371
Epoch 86/100
47/47 [==============================] - 0s 3ms/step - loss: -77.7431 - accuracy:
0.5344 - val_loss: -124.6676 - val_accuracy: 0.4874
Epoch 87/100
47/47 [==============================] - 0s 4ms/step - loss: -84.1267 - accuracy:
0.5506 - val_loss: -27.7219 - val_accuracy: 0.5943
Epoch 88/100
47/47 [==============================] - 0s 4ms/step - loss: -86.5639 - accuracy:
0.5506 - val_loss: -124.0542 - val_accuracy: 0.4906
Epoch 89/100
47/47 [==============================] - 0s 4ms/step - loss: -97.3375 - accuracy:
0.5533 - val_loss: -120.2535 - val_accuracy: 0.4403
Epoch 90/100
47/47 [==============================] - 0s 3ms/step - loss: -86.7052 - accuracy:
0.5344 - val_loss: -50.2680 - val_accuracy: 0.5786
Epoch 91/100
47/47 [==============================] - 0s 3ms/step - loss: -85.9945 - accuracy:
0.5385 - val_loss: -39.0791 - val_accuracy: 0.6792
Epoch 92/100
47/47 [==============================] - 0s 4ms/step - loss: -89.6942 - accuracy:
0.5466 - val_loss: -102.0823 - val_accuracy: 0.5346
Epoch 93/100
47/47 [==============================] - 0s 4ms/step - loss: -85.7191 - accuracy:
0.5547 - val_loss: -176.4807 - val_accuracy: 0.4591
Epoch 94/100
47/47 [==============================] - 0s 3ms/step - loss: -97.1916 - accuracy:
0.5439 - val_loss: -156.8604 - val_accuracy: 0.4811
Epoch 95/100
47/47 [==============================] - 0s 4ms/step - loss: -96.4436 - accuracy:
0.5452 - val_loss: -159.1734 - val_accuracy: 0.4717
Epoch 96/100
47/47 [==============================] - 0s 4ms/step - loss: -92.4171 - accuracy:
0.5331 - val_loss: -134.0214 - val_accuracy: 0.4811
Epoch 97/100
47/47 [==============================] - 0s 4ms/step - loss: -103.1512 - accuracy:
0.5547 - val_loss: -50.2631 - val_accuracy: 0.6132
Epoch 98/100
47/47 [==============================] - 0s 3ms/step - loss: -89.5648 - accuracy:
0.5304 - val_loss: -52.7098 - val_accuracy: 0.5597
Epoch 99/100
47/47 [==============================] - 0s 3ms/step - loss: -111.4318 - accuracy:
0.5466 - val_loss: -50.3987 - val_accuracy: 0.4088
Epoch 100/100
47/47 [==============================] - 0s 3ms/step - loss: -108.6408 - accuracy:
0.5506 - val_loss: -170.4726 - val_accuracy: 0.5377

In [59]: ann_model.summary()

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 26/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense (Dense) (None, 32) 256

batch_normalization (BatchN (None, 32) 128

ormalization)

dropout (Dropout) (None, 32) 0

dense_1 (Dense) (None, 64) 2112

batch_normalization_1 (Batc (None, 64) 256

hNormalization)

dropout_1 (Dropout) (None, 64) 0

dense_2 (Dense) (None, 1) 65

=================================================================
Total params: 2,817
Trainable params: 2,625
Non-trainable params: 192
_________________________________________________________________

In [60]: loss_plot = pd.DataFrame(ann_model.history.history)

loss_plot.plot()

Out[60]: <AxesSubplot:>

In [61]: #now testing for Test data

y_pred = ann_model.predict(x_test)
y_pred = (y_pred>0.5)
acc_test_ann1 = round(metrics.accuracy_score(y_test,y_pred) * 100, 2)
acc_test_ann1

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 27/28
4/4/23, 12:48 PM milk-quality-prediction-using-machine-learning

10/10 [==============================] - 0s 1ms/step

Out[61]: 53.77

In [62]: from sklearn.metrics import confusion_matrix

from sklearn.metrics import accuracy_score
cm = confusion_matrix(y_test,y_pred)
acc_test_ann1 = accuracy_score(y_test,y_pred)
print(cm)
print('score is:',acc_test_ann1)

[[ 65 51 0]
[ 10 106 0]
[ 0 86 0]]
score is: 0.5377358490566038

In [63]: print(classification_report(y_test,y_pred))

precision recall f1-score support

0 0.87 0.56 0.68 116

1 0.44 0.91 0.59 116
2 0.00 0.00 0.00 86

accuracy 0.54 318

macro avg 0.43 0.49 0.42 318
weighted avg 0.48 0.54 0.46 318

Prediction
In [64]: print(xgb.predict(x_test))

[2 0 1 0 2 1 0 2 2 1 0 2 1 1 2 1 2 2 2 0 2 0 1 2 2 0 2 2 0 1 2 0 0 1 2 2 0
1 2 0 2 1 2 0 1 0 2 0 1 0 2 1 1 1 0 1 1 0 1 0 1 0 1 2 0 1 0 1 1 2 2 1 1 2
1 1 1 0 2 1 0 1 1 0 1 1 2 0 2 0 1 2 1 0 2 1 0 2 1 1 2 1 0 1 2 2 2 2 0 0 2
2 2 2 0 1 0 2 1 2 1 1 0 2 2 0 0 0 1 2 0 1 0 1 2 1 1 1 1 0 1 2 0 0 1 2 0 0
0 2 0 1 1 0 2 2 1 1 1 1 1 1 0 1 1 2 2 0 2 1 0 0 1 2 2 0 2 2 2 1 1 0 0 0 0
2 1 2 0 0 2 0 0 0 1 0 1 1 2 1 0 2 1 0 1 1 0 1 1 0 2 1 1 1 1 1 2 1 2 0 2 2
1 2 1 0 0 1 2 2 2 0 0 2 0 1 0 0 0 2 0 2 0 1 2 0 1 1 0 0 1 0 2 0 0 0 2 0 2
0 0 1 1 0 1 1 0 1 1 1 1 1 0 0 0 2 1 0 0 1 1 0 2 2 0 1 2 1 1 1 0 1 0 0 0 0
0 0 0 0 1 1 0 1 1 1 1 0 1 1 1 0 0 0 2 0 2 0]

In [ ]:

localhost:8888/nbconvert/html/Downloads/milk-quality-prediction-using-machine-learning.ipynb?download=false 28/28