from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from keras.models import Sequential
from keras.layers import Dense
import pandas as pd
data = pd.read_csv(r"Book_updated.csv")
target = data["Prakruti type"]
train = data.drop(['Prakruti type'],axis = 1)
classes = train.columns
encoders = []
unique_output=[]
for col in train.columns:
    le = LabelEncoder()
    unique_output.append((train[col].unique()).tolist())
    train[col] = le.fit_transform(train[col])
    encoders.append(le)
target2 = pd.get_dummies(target)
model = Sequential([
    Dense(64,activation='relu',input_shape=(18,)),
    Dense(32,activation='relu'),
    Dense(7,activation='softmax')
])
model.compile(optimizer='adam',metrics='categorical_crossentropy',loss='mean_squared_error')
x_tr,x_te,y_tr,y_te=train_test_split(train,target2,random_state=123,test_size=0.2)
model.fit(x_tr,y_tr,epochs = 1000,batch_size=12,verbose= 1)