Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 8284797

Browse files
Merge pull request avinashkranjan#1878 from Killer2OP/stock
[ADDED]: Stock-Price-Prediction
2 parents 34d6426 + 94d1fd3 commit 8284797

File tree

4 files changed

+335
-0
lines changed

4 files changed

+335
-0
lines changed

‎Stock-Price-Prediction/README.md

Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
2+
<h2 align=center> 📑 Stock Price Prediction </h2>
3+
4+
Machine learning has significant applications in the stock price prediction. In this machine learning project, we will be talking about predicting the returns on stocks. This is a very complex task and has uncertainties. We will develop this project into two parts:
5+
6+
First, we will predict stock price using the LSTM neural network.<br>
7+
Then we will build a dashboard using Plotly dash for stock analysis.
8+
9+
<h3>Datasets</h3>
10+
11+
To build the stock price prediction model, we will use the NSE TATA GLOBAL dataset. This is a dataset of Tata Beverages from Tata Global Beverages Limited, National Stock Exchange of India: Tata Global Dataset
12+
To develop the dashboard for stock analysis we will use another stock dataset with multiple stocks like Apple, Microsoft, Facebook: Stocks Dataset
13+
14+
![Alt text](Stock-Price-Prediction-project-dashboard-2.gif)
1.1 MB
Loading[フレーム]

‎Stock-Price-Prediction/stock_app.py

Lines changed: 236 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,236 @@
1+
import dash
2+
import dash_core_components as dcc
3+
import dash_html_components as html
4+
import pandas as pd
5+
import plotly.graph_objs as go
6+
from dash.dependencies import Input, Output
7+
from keras.models import load_model
8+
from sklearn.preprocessing import MinMaxScaler
9+
import numpy as np
10+
11+
12+
app = dash.Dash()
13+
server = app.server
14+
15+
scaler=MinMaxScaler(feature_range=(0,1))
16+
17+
18+
19+
df_nse = pd.read_csv("./NSE-TATA.csv")
20+
21+
df_nse["Date"]=pd.to_datetime(df_nse.Date,format="%Y-%m-%d")
22+
df_nse.index=df_nse['Date']
23+
24+
25+
data=df_nse.sort_index(ascending=True,axis=0)
26+
new_data=pd.DataFrame(index=range(0,len(df_nse)),columns=['Date','Close'])
27+
28+
for i in range(0,len(data)):
29+
new_data["Date"][i]=data['Date'][i]
30+
new_data["Close"][i]=data["Close"][i]
31+
32+
new_data.index=new_data.Date
33+
new_data.drop("Date",axis=1,inplace=True)
34+
35+
dataset=new_data.values
36+
37+
train=dataset[0:987,:]
38+
valid=dataset[987:,:]
39+
40+
scaler=MinMaxScaler(feature_range=(0,1))
41+
scaled_data=scaler.fit_transform(dataset)
42+
43+
x_train,y_train=[],[]
44+
45+
for i in range(60,len(train)):
46+
x_train.append(scaled_data[i-60:i,0])
47+
y_train.append(scaled_data[i,0])
48+
49+
x_train,y_train=np.array(x_train),np.array(y_train)
50+
51+
x_train=np.reshape(x_train,(x_train.shape[0],x_train.shape[1],1))
52+
53+
model=load_model("saved_model.h5")
54+
55+
inputs=new_data[len(new_data)-len(valid)-60:].values
56+
inputs=inputs.reshape(-1,1)
57+
inputs=scaler.transform(inputs)
58+
59+
X_test=[]
60+
for i in range(60,inputs.shape[0]):
61+
X_test.append(inputs[i-60:i,0])
62+
X_test=np.array(X_test)
63+
64+
X_test=np.reshape(X_test,(X_test.shape[0],X_test.shape[1],1))
65+
closing_price=model.predict(X_test)
66+
closing_price=scaler.inverse_transform(closing_price)
67+
68+
train=new_data[:987]
69+
valid=new_data[987:]
70+
valid['Predictions']=closing_price
71+
72+
73+
74+
df= pd.read_csv("./stock_data.csv")
75+
76+
app.layout = html.Div([
77+
78+
html.H1("Stock Price Analysis Dashboard", style={"textAlign": "center"}),
79+
80+
dcc.Tabs(id="tabs", children=[
81+
82+
dcc.Tab(label='NSE-TATAGLOBAL Stock Data',children=[
83+
html.Div([
84+
html.H2("Actual closing price",style={"textAlign": "center"}),
85+
dcc.Graph(
86+
id="Actual Data",
87+
figure={
88+
"data":[
89+
go.Scatter(
90+
x=train.index,
91+
y=valid["Close"],
92+
mode='markers'
93+
)
94+
95+
],
96+
"layout":go.Layout(
97+
title='scatter plot',
98+
xaxis={'title':'Date'},
99+
yaxis={'title':'Closing Rate'}
100+
)
101+
}
102+
103+
),
104+
html.H2("LSTM Predicted closing price",style={"textAlign": "center"}),
105+
dcc.Graph(
106+
id="Predicted Data",
107+
figure={
108+
"data":[
109+
go.Scatter(
110+
x=valid.index,
111+
y=valid["Predictions"],
112+
mode='markers'
113+
)
114+
115+
],
116+
"layout":go.Layout(
117+
title='scatter plot',
118+
xaxis={'title':'Date'},
119+
yaxis={'title':'Closing Rate'}
120+
)
121+
}
122+
123+
)
124+
])
125+
126+
127+
]),
128+
dcc.Tab(label='Facebook Stock Data', children=[
129+
html.Div([
130+
html.H1("Stocks High vs Lows",
131+
style={'textAlign': 'center'}),
132+
133+
dcc.Dropdown(id='my-dropdown',
134+
options=[{'label': 'Tesla', 'value': 'TSLA'},
135+
{'label': 'Apple','value': 'AAPL'},
136+
{'label': 'Facebook', 'value': 'FB'},
137+
{'label': 'Microsoft','value': 'MSFT'}],
138+
multi=True,value=['FB'],
139+
style={"display": "block", "margin-left": "auto",
140+
"margin-right": "auto", "width": "60%"}),
141+
dcc.Graph(id='highlow'),
142+
html.H1("Stocks Market Volume", style={'textAlign': 'center'}),
143+
144+
dcc.Dropdown(id='my-dropdown2',
145+
options=[{'label': 'Tesla', 'value': 'TSLA'},
146+
{'label': 'Apple','value': 'AAPL'},
147+
{'label': 'Facebook', 'value': 'FB'},
148+
{'label': 'Microsoft','value': 'MSFT'}],
149+
multi=True,value=['FB'],
150+
style={"display": "block", "margin-left": "auto",
151+
"margin-right": "auto", "width": "60%"}),
152+
dcc.Graph(id='volume')
153+
], className="container"),
154+
])
155+
156+
157+
])
158+
])
159+
160+
161+
162+
163+
164+
165+
166+
@app.callback(Output('highlow', 'figure'),
167+
[Input('my-dropdown', 'value')])
168+
def update_graph(selected_dropdown):
169+
dropdown = {"TSLA": "Tesla","AAPL": "Apple","FB": "Facebook","MSFT": "Microsoft",}
170+
trace1 = []
171+
trace2 = []
172+
for stock in selected_dropdown:
173+
trace1.append(
174+
go.Scatter(x=df[df["Stock"] == stock]["Date"],
175+
y=df[df["Stock"] == stock]["High"],
176+
mode='lines', opacity=0.7,
177+
name=f'High {dropdown[stock]}',textposition='bottom center'))
178+
trace2.append(
179+
go.Scatter(x=df[df["Stock"] == stock]["Date"],
180+
y=df[df["Stock"] == stock]["Low"],
181+
mode='lines', opacity=0.6,
182+
name=f'Low {dropdown[stock]}',textposition='bottom center'))
183+
traces = [trace1, trace2]
184+
data = [val for sublist in traces for val in sublist]
185+
figure = {'data': data,
186+
'layout': go.Layout(colorway=["#5E0DAC", '#FF4F00', '#375CB1',
187+
'#FF7400', '#FFF400', '#FF0056'],
188+
height=600,
189+
title=f"High and Low Prices for {', '.join(str(dropdown[i]) for i in selected_dropdown)} Over Time",
190+
xaxis={"title":"Date",
191+
'rangeselector': {'buttons': list([{'count': 1, 'label': '1M',
192+
'step': 'month',
193+
'stepmode': 'backward'},
194+
{'count': 6, 'label': '6M',
195+
'step': 'month',
196+
'stepmode': 'backward'},
197+
{'step': 'all'}])},
198+
'rangeslider': {'visible': True}, 'type': 'date'},
199+
yaxis={"title":"Price (USD)"})}
200+
return figure
201+
202+
203+
@app.callback(Output('volume', 'figure'),
204+
[Input('my-dropdown2', 'value')])
205+
def update_graph(selected_dropdown_value):
206+
dropdown = {"TSLA": "Tesla","AAPL": "Apple","FB": "Facebook","MSFT": "Microsoft",}
207+
trace1 = []
208+
for stock in selected_dropdown_value:
209+
trace1.append(
210+
go.Scatter(x=df[df["Stock"] == stock]["Date"],
211+
y=df[df["Stock"] == stock]["Volume"],
212+
mode='lines', opacity=0.7,
213+
name=f'Volume {dropdown[stock]}', textposition='bottom center'))
214+
traces = [trace1]
215+
data = [val for sublist in traces for val in sublist]
216+
figure = {'data': data,
217+
'layout': go.Layout(colorway=["#5E0DAC", '#FF4F00', '#375CB1',
218+
'#FF7400', '#FFF400', '#FF0056'],
219+
height=600,
220+
title=f"Market Volume for {', '.join(str(dropdown[i]) for i in selected_dropdown_value)} Over Time",
221+
xaxis={"title":"Date",
222+
'rangeselector': {'buttons': list([{'count': 1, 'label': '1M',
223+
'step': 'month',
224+
'stepmode': 'backward'},
225+
{'count': 6, 'label': '6M',
226+
'step': 'month',
227+
'stepmode': 'backward'},
228+
{'step': 'all'}])},
229+
'rangeslider': {'visible': True}, 'type': 'date'},
230+
yaxis={"title":"Transactions Volume"})}
231+
return figure
232+
233+
234+
235+
if __name__=='__main__':
236+
app.run_server(debug=True)

‎Stock-Price-Prediction/stock_pred.py

Lines changed: 85 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,85 @@
1+
import pandas as pd
2+
import numpy as np
3+
4+
import matplotlib.pyplot as plt
5+
%matplotlib inline
6+
7+
from matplotlib.pylab import rcParams
8+
rcParams['figure.figsize']=20,10
9+
10+
from sklearn.preprocessing import MinMaxScaler
11+
scaler=MinMaxScaler(feature_range=(0,1))
12+
13+
df=pd.read_csv("NSE-TATA.csv")
14+
df.head()
15+
16+
df["Date"]=pd.to_datetime(df.Date,format="%Y-%m-%d")
17+
df.index=df['Date']
18+
19+
plt.figure(figsize=(16,8))
20+
plt.plot(df["Close"],label='Close Price history')
21+
22+
from keras.models import Sequential
23+
from keras.layers import LSTM,Dropout,Dense
24+
25+
data=df.sort_index(ascending=True,axis=0)
26+
new_dataset=pd.DataFrame(index=range(0,len(df)),columns=['Date','Close'])
27+
28+
for i in range(0,len(data)):
29+
new_dataset["Date"][i]=data['Date'][i]
30+
new_dataset["Close"][i]=data["Close"][i]
31+
32+
33+
new_dataset.index=new_dataset.Date
34+
new_dataset.drop("Date",axis=1,inplace=True)
35+
36+
final_dataset=new_dataset.values
37+
38+
train_data=final_dataset[0:987,:]
39+
valid_data=final_dataset[987:,:]
40+
41+
scaler=MinMaxScaler(feature_range=(0,1))
42+
scaled_data=scaler.fit_transform(final_dataset)
43+
44+
x_train_data,y_train_data=[],[]
45+
46+
for i in range(60,len(train_data)):
47+
x_train_data.append(scaled_data[i-60:i,0])
48+
y_train_data.append(scaled_data[i,0])
49+
50+
x_train_data,y_train_data=np.array(x_train_data),np.array(y_train_data)
51+
52+
x_train_data=np.reshape(x_train_data,(x_train_data.shape[0],x_train_data.shape[1],1))
53+
54+
lstm_model=Sequential()
55+
lstm_model.add(LSTM(units=50,return_sequences=True,input_shape=(x_train_data.shape[1],1)))
56+
lstm_model.add(LSTM(units=50))
57+
lstm_model.add(Dense(1))
58+
59+
60+
61+
62+
lstm_model.compile(loss='mean_squared_error',optimizer='adam')
63+
lstm_model.fit(x_train_data,y_train_data,epochs=1,batch_size=1,verbose=2)
64+
65+
inputs_data=new_dataset[len(new_dataset)-len(valid_data)-60:].values
66+
inputs_data=inputs_data.reshape(-1,1)
67+
inputs_data=scaler.transform(inputs_data)
68+
69+
70+
X_test=[]
71+
for i in range(60,inputs_data.shape[0]):
72+
X_test.append(inputs_data[i-60:i,0])
73+
X_test=np.array(X_test)
74+
75+
X_test=np.reshape(X_test,(X_test.shape[0],X_test.shape[1],1))
76+
closing_price=model.predict(X_test)
77+
closing_price=scaler.inverse_transform(closing_price)
78+
79+
lstm_model.save("saved_lstm_model.h5")
80+
81+
train_data=new_dataset[:987]
82+
valid_data=new_dataset[987:]
83+
valid_data['Predictions']=prediction_closing
84+
plt.plot(train_data["Close"])
85+
plt.plot(valid_data[['Close',"Predictions"]])

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /