Manu Siddhartha sid321axn

## gist:9e540d9a028978205cbbd1fcb2518539
from autoviz.AutoViz_Class import AutoViz_Class
AV = AutoViz_Class()
df = AV.AutoViz('chronic_cleaned.csv',depVar='class')

## gist:efe9bf65c17c1f5f155941e419cef6d0
sv_report = sv.analyze(df,"class")
#display the report
sv_report.show_html('eda.html')

## gist:a8913d33e609273625c68a34655f24f2
import sweetviz as sv

sv_report = sv.analyze(df)
#display the report
sv_report.show_html('eda.html')

## gist:f63a17f60fa0df3f39b63f10ece91b31
from pandas_profiling import ProfileReport

profile=ProfileReport(df)
profile.to_file('report.html')

## gist:be47865c9a86db6ee882e080997cb371
profile=ProfileReport(df,title='Chronic Kidney disease data profile',html={'style':{'full_width':True}})

profile

## gist:5d45823cb73690e0921e874f98734c08
df=pd.read_csv('chronic_cleaned.csv')

## gist:8d2842a35e89fd6c8de4da827738993a
import pandas as pd
from pandas_profiling import ProfileReport

## inverse transform
#transforming predicted values
inv_yhat = sc.inverse_transform(y_pred_test_gru)

#transforming actual values of test set
inv_ytest = sc.inverse_transform(y_test)

## gru compile
model_gru.compile(loss=tf.keras.metrics.mean_squared_error,
              metrics=[tf.keras.metrics.RootMeanSquaredError(name='rmse')], optimizer='adam')

early_stop = EarlyStopping(monitor='loss', patience=10, verbose=1)

history_model_gru = model_gru.fit(X_tr_t, y_train, epochs=100, batch_size=1, verbose=1, shuffle=False, callbacks=[early_stop])

## GRU
from keras.layers import GRU
K.clear_session()
model_gru = Sequential()
model_gru.add(GRU(7, input_shape=(1, X_train.shape[1]), activation='linear', kernel_initializer='lecun_uniform', return_sequences=False))
model_gru.add(Dense(1))
model_gru.summary()
	from autoviz.AutoViz_Class import AutoViz_Class
	AV = AutoViz_Class()
	df = AV.AutoViz('chronic_cleaned.csv',depVar='class')
	sv_report = sv.analyze(df,"class")
	#display the report
	sv_report.show_html('eda.html')
	import sweetviz as sv

	sv_report = sv.analyze(df)
	#display the report
	sv_report.show_html('eda.html')
	from pandas_profiling import ProfileReport

	profile=ProfileReport(df)
	profile.to_file('report.html')
	profile=ProfileReport(df,title='Chronic Kidney disease data profile',html={'style':{'full_width':True}})

	profile
	import pandas as pd
	from pandas_profiling import ProfileReport
	#transforming predicted values
	inv_yhat = sc.inverse_transform(y_pred_test_gru)

	#transforming actual values of test set
	inv_ytest = sc.inverse_transform(y_test)
	model_gru.compile(loss=tf.keras.metrics.mean_squared_error,
	metrics=[tf.keras.metrics.RootMeanSquaredError(name='rmse')], optimizer='adam')

	early_stop = EarlyStopping(monitor='loss', patience=10, verbose=1)

	history_model_gru = model_gru.fit(X_tr_t, y_train, epochs=100, batch_size=1, verbose=1, shuffle=False, callbacks=[early_stop])
	from keras.layers import GRU
	K.clear_session()
	model_gru = Sequential()
	model_gru.add(GRU(7, input_shape=(1, X_train.shape[1]), activation='linear', kernel_initializer='lecun_uniform', return_sequences=False))
	model_gru.add(Dense(1))
	model_gru.summary()