import pandas as pd
import numpy as np
import time
import requests
import matplotlib.pyplot as plt

from sklearn.preprocessing import MinMaxScaler

import tensorflow as tf
from keras.models import Sequential
from keras.layers import Dense,Dropout,Bidirectional
from keras.layers import LSTM

data=requests.get('https://api.twelvedata.com/time_series?symbol=BTC/INR&timezone=Asia/Kolkata&start_date=2023-03-01 00:00:00&end_date=2023-03-20 00:00:00&order=ASC&interval=5min&outputsize=5000&apikey=e76157c75c3a42649e168c5c206e88ca').json()
#print(data)
data_final=pd.DataFrame(data['values'])
#print(data_final)
scaler=MinMaxScaler(feature_range=(0,1))
scaled_data=scaler.fit_transform(data_final['close'].values.reshape(-1,1))
timeinterval=24
prediction=1

x_train=[]
y_train=[]

for i in range(timeinterval,len(scaled_data)-prediction):
  x_train.append(scaled_data[i-timeinterval:i,0])
  y_train.append(scaled_data[i+prediction,0])

x_train=np.array(x_train)
y_train=np.array(y_train)
x_train=np.reshape(x_train,(x_train.shape[0],x_train.shape[1],1))

model=Sequential()

model.add(Bidirectional(LSTM(256,return_sequences=True,input_shape=(x_train.shape[1],1),activation='relu')))
model.add(Dropout(0.3))
model.add(Bidirectional(LSTM(128,return_sequences=True,activation='tanh')))
model.add(Dropout(0.2))
model.add(Bidirectional(LSTM(64,activation='relu')))
model.add(Dropout(0.1))
model.add(Dense(1,activation='sigmoid'))
model.compile(loss='mean_squared_error',optimizer='adam',metrics=['accuracy'])
model.fit(x_train,y_train,epochs=15,batch_size=100)
predict=[]
exchange=[]

i=0
while(i<6):
  testapi='https://api.twelvedata.com/time_series?symbol=BTC/INR&interval=5min&outputsize=3000&timezone=Asia/Kolkata&order=ASC&apikey=e76157c75c3a42649e168c5c206e88ca'
  testdata=requests.get(testapi).json()
  testdatafinal=pd.DataFrame(testdata['values'])

  bitcoinprice=pd.to_numeric(testdatafinal['close'],errors='coerce').values
  testinputs=testdatafinal['close'].values
  testinputs=testinputs.reshape(-1,1)
  modelinputs=scaler.fit_transform(testinputs)

  x_test=[]
  for x in range(timeinterval,len(modelinputs)):
    x_test.append(modelinputs[x-timeinterval:x,0])

  x_test=np.array(x_test)
  x_test=np.reshape(x_test,(x_test.shape[0],x_test.shape[1],1))
  prediction_price=model.predict(x_test)
  prediction_price=scaler.inverse_transform(prediction_price)

  """plt.plot(bitcoinprice,label='Bitcoin Prices')
  plt.plot(prediction_price,label='Predicted Prices')
  plt.title('Predicting Bitcoin Price')
  plt.xlabel('5min Interval')
  plt.ylabel('Price')
  plt.legend()
  plt.show()"""
  exchangeapi='https://api.twelvedata.com/exchange_rate?symbol=BTC/INR&timezone=Asia/Kolkata&apikey=e76157c75c3a42649e168c5c206e88ca'
  exchangedata=requests.get(exchangeapi).json()
  print(exchangedata)

  exchangefinal=exchangedata['rate']
  exchange.append(exchangefinal)


  lastdata=modelinputs[len(modelinputs)+1-timeinterval:len(modelinputs)+1,0]
  lastdata=np.array(lastdata)
  lastdata=np.reshape(lastdata,(1,lastdata.shape[0],1))
  prediction=model.predict(lastdata)
  prediction=scaler.inverse_transform(prediction)
  predict.append(prediction[0][0])
  print(prediction)
  i+=1
  print(i)
  time.sleep(300)

  """testapi='https://api.twelvedata.com/time_series?symbol=BTC/INR&interval=5min&outputsize=3000&apikey=e76157c75c3a42649e168c5c206e88ca'
  testdata=requests.get(testapi).json()
  testdatafinal=pd.DataFrame(testdata['values'])

  bitcoinprice=pd.to_numeric(testdatafinal['close'],errors='coerce').values
  testinputs=testdatafinal['close'].values
  testinputs=testinputs.reshape(-1,1)
  modelinputs=scaler.fit_transform(testinputs)

  x_test=[]
  for x in range(timeinterval,len(modelinputs)):
    x_test.append(modelinputs[x-timeinterval:x,0])

  x_test=np.array(x_test)
  x_test=np.reshape(x_test,(x_test.shape[0],x_test.shape[1],1))
  prediction_price=model.predict(x_test)
  prediction_price=scaler.inverse_transform(prediction_price)

plt.plot(bitcoinprice,label='Bitcoin Prices')
plt.plot(prediction_price,label='Predicted Prices')
plt.title('Predicting Bitcoin Price')
plt.xlabel('5min Interval')
plt.ylabel('Price')
plt.legend()
plt.show()
predict=[]
exchange=[]
exchangeapi='https://api.twelvedata.com/exchange_rate?symbol=BTC/INR&timezone=Asia/Kolkata&apikey=e76157c75c3a42649e168c5c206e8'
exchangedata=requests.get(exchangeapi).json()
exchangefinal=pd.DataFrame(exchangedata['rate'])
exchange.append(exchangefinal)




lastdata=modelinputs[len(modelinputs)+1-timeinterval:len(modelinputs)+1,0]
lastdata=np.array(lastdata)
lastdata=np.reshape(lastdata,(1,lastdata.shape[0],1))
prediction=model.predict(lastdata)
prediction=scaler.inverse_transform(prediction)
predict.append(prediction)
print(prediction)"""



exchange1api='https://api.twelvedata.com/exchange_rate?symbol=BTC/INR&timezone=Asia/Kolkata&apikey=e76157c75c3a42649e168c5c206e88ca'
exchange1data=requests.get(exchange1api).json()
exchange1final=exchange1data['rate']
exchange.append(exchange1final)
exchange.pop(0)
print(exchange)
print(predict)

plt.plot(exchange,label='Bitcoin Price')
plt.plot(predict,label='Predicted Prices')
plt.title('Predicting Bitcoin Price')
plt.xlabel('5min Interval')
plt.ylabel('Price')
plt.legend()
plt.show()