Update app.py

app.py

@@ -2,8 +2,7 @@ import datetime
 import gradio as gr
 import pandas as pd
 import yfinance as yf
-import seaborn as sns;
-
+import seaborn as sns
 sns.set()
 import matplotlib.pyplot as plt
 import plotly.graph_objects as go
@@ -18,6 +17,7 @@ from dateutil.relativedelta import relativedelta
 
 index_options = ['FTSE 100(UK)', 'NASDAQ(USA)', 'CAC 40(FRANCE)']
 ticker_dict = {'FTSE 100(UK)': 'FTSE 100', 'NASDAQ(USA)': 'NASDAQ 100', 'CAC 40(FRANCE)': 'CAC 40'}
+time_intervals = ['1d', '1m', '5m', '15m', '60m']
 
 global START_DATE, END_DATE
 
@@ -28,23 +28,18 @@ demo = gr.Blocks()
 stock_names = []
 
 with demo:
-    d1 = gr.Dropdown(index_options, label='Please select Index...',
-                     info='Will be adding more indices later on',
-                     interactive=True)
-
-    d2 = gr.Dropdown([])  # for specific stocks
-
-
-    # d3 = gr.Dropdown(['General News'])
+    d1 = gr.Dropdown(index_options, label='Please select Index...', info='Will be adding more indices later on', interactive=True)
+    d2 = gr.Dropdown([], label='Please Select Stock from your selected index', interactive=True)
+    d3 = gr.Dropdown(time_intervals, label='Select Time Interval', value='1d', interactive=True)
+    d4 = gr.Radio(['Line Graph', 'Candlestick Graph'], label='Select Graph Type', value='Line Graph', interactive=True)
+    d5 = gr.Dropdown(['ARIMA', 'Prophet', 'LSTM'], label='Select Forecasting Method', value='ARIMA', interactive=True)
 
     def forecast_series(series, model="ARIMA", forecast_period=7):
-
         predictions = list()
         if series.shape[1] > 1:
             series = series['Close'].values.tolist()
-        plt.show()
+        
         if model == "ARIMA":
-            ## Do grid search here --> Custom for all stocks
             for i in range(forecast_period):
                 model = ARIMA(series, order=(5, 1, 0))
                 model_fit = model.fit()
@@ -52,74 +47,69 @@ with demo:
                 yhat = output[0]
                 predictions.append(yhat)
                 series.append(yhat)
+        elif model == "Prophet":
+            # Implement Prophet forecasting method
+            pass
+        elif model == "LSTM":
+            # Implement LSTM forecasting method
+            pass
 
         return predictions
 
-
     def is_business_day(a_date):
         return a_date.weekday() < 5
 
-
     def get_stocks_from_index(idx):
         stock_data = PyTickerSymbols()
-        # indices = stock_data.get_all_indices()
         index = ticker_dict[idx]
-        stock_data = PyTickerSymbols()
-
-        # returns 2d list with the following information
-        # 'name', 'symbol', 'country', 'indices', 'industries', 'symbols', 'metadata', 'isins', 'akas'
-        stocks = list(stock_data.get_stocks_by_index(index))  ##converting filter object to list
-        stock_names = []
-        for stock in stocks:
-            stock_names.append(stock['name'] + ':' + stock['symbol'])
-        d2 = gr.Dropdown(choices=stock_names, label='Please Select Stock from your selected index', interactive=True)
-        return d2
-
+        stocks = list(stock_data.get_stocks_by_index(index))
+        stock_names = [f"{stock['name']}:{stock['symbol']}" for stock in stocks]
+        return gr.Dropdown(choices=stock_names, label='Please Select Stock from your selected index', interactive=True)
 
     d1.input(get_stocks_from_index, d1, d2)
-    out = gr.Plot(every=10)
-
-
-    def get_stock_graph(idx, stock):
-
-        stock_name = stock.split(":")[0]
-        ticker_name = stock.split(":")[1]
 
+    def get_stock_graph(idx, stock, interval, graph_type, forecast_method):
+        stock_name, ticker_name = stock.split(":")
+        
         if ticker_dict[idx] == 'FTSE 100':
-            if ticker_name[-1] == '.':
-                ticker_name += 'L'
-            else:
-                ticker_name += '.L'
+            ticker_name += '.L' if ticker_name[-1] != '.' else 'L'
         elif ticker_dict[idx] == 'CAC 40':
             ticker_name += '.PA'
 
-        ## Can also download lower interval data apparently using line below
-        # data = yf.download(tickers="MSFT", period="5d", interval="1m")
-        series = yf.download(tickers=ticker_name, start=START_DATE, end=END_DATE)  # stock.split(":")[1]
+        series = yf.download(tickers=ticker_name, start=START_DATE, end=END_DATE, interval=interval)
         series = series.reset_index()
 
-        predictions = forecast_series(series)
+        predictions = forecast_series(series, model=forecast_method)
 
         last_date = pd.to_datetime(series['Date'].values[-1])
-        forecast_week = []
-
-        while len(forecast_week) != FORECAST_PERIOD:
-            if is_business_day(last_date):
-                forecast_week.append(last_date)
-            last_date += timedelta(days=1)
+        forecast_week = [last_date + timedelta(days=i) for i in range(1, FORECAST_PERIOD + 1) if is_business_day(last_date + timedelta(days=i))]
 
         forecast = pd.DataFrame({"Date": forecast_week, "Forecast": predictions})
 
-        fig = plt.figure(figsize=(14, 5))
-        sns.set_style("ticks")
-        sns.lineplot(data=series, x="Date", y="Close", color="firebrick")
-        sns.lineplot(data=forecast, x="Date", y="Forecast", color="blue")
-        sns.despine()
+        if graph_type == 'Line Graph':
+            fig = go.Figure()
+            fig.add_trace(go.Scatter(x=series['Date'], y=series['Close'], mode='lines', name='Historical'))
+            fig.add_trace(go.Scatter(x=forecast['Date'], y=forecast['Forecast'], mode='lines', name='Forecast'))
+        else:  # Candlestick Graph
+            fig = go.Figure(data=[go.Candlestick(x=series['Date'],
+                                                 open=series['Open'],
+                                                 high=series['High'],
+                                                 low=series['Low'],
+                                                 close=series['Close'],
+                                                 name='Historical')])
+            fig.add_trace(go.Scatter(x=forecast['Date'], y=forecast['Forecast'], mode='lines', name='Forecast'))
+
+        fig.update_layout(title=f"Stock Price of {stock_name}",
+                          xaxis_title="Date",
+                          yaxis_title="Price")
 
-        plt.title("Stock Price of {}".format(stock_name), size='x-large', color='blue')  # stock.split(":")[0]
-        text = "Your stock is:" + str(stock)
         return fig
 
+    out = gr.Plot()
+    inputs = [d1, d2, d3, d4, d5]
+    d2.input(get_stock_graph, inputs, out)
+    d3.input(get_stock_graph, inputs, out)
+    d4.input(get_stock_graph, inputs, out)
+    d5.input(get_stock_graph, inputs, out)
 
-    d2.input(get_stock_graph, [d1, d2], out)
 demo.launch()