Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -7,7 +7,7 @@ import gradio as gr
|
|
7 |
# Set FRED API Key from environment variable
|
8 |
FRED_API_KEY = os.getenv('FRED_API_KEY')
|
9 |
|
10 |
-
#
|
11 |
series_options = {
|
12 |
"UNRATE": "Unemployment Rate",
|
13 |
"GDP": "Gross Domestic Product",
|
@@ -17,13 +17,12 @@ series_options = {
|
|
17 |
"M1SL": "M1 Money Supply",
|
18 |
"M2SL": "M2 Money Supply",
|
19 |
"M3SL": "M3 Money Supply",
|
20 |
-
# Add more series as needed
|
21 |
"HOUST": "Housing Starts",
|
22 |
"PCE": "Personal Consumption Expenditures",
|
23 |
-
"BAA10YM": "Moody's Baa Corporate Bond Yield Spread"
|
24 |
}
|
25 |
|
26 |
-
# Function to fetch data from FRED API
|
27 |
def fetch_fred_data(series_ids):
|
28 |
"""
|
29 |
Fetches data for a list of FRED series IDs.
|
@@ -33,12 +32,16 @@ def fetch_fred_data(series_ids):
|
|
33 |
for series_id in series_ids:
|
34 |
response = requests.get(
|
35 |
f'https://api.stlouisfed.org/fred/series/observations',
|
36 |
-
params={
|
|
|
|
|
|
|
|
|
37 |
)
|
38 |
if response.status_code == 200:
|
39 |
observations = response.json().get('observations', [])
|
40 |
dates = [obs['date'] for obs in observations]
|
41 |
-
#
|
42 |
values = [
|
43 |
float(obs['value']) if obs['value'].replace('.', '', 1).isdigit() else float('nan')
|
44 |
for obs in observations
|
@@ -55,22 +58,31 @@ def standardize_data(df):
|
|
55 |
"""
|
56 |
return (df - df.mean()) / df.std()
|
57 |
|
58 |
-
# Function to create 3D correlation matrix
|
59 |
def create_3d_correlation_matrix(df):
|
|
|
|
|
|
|
|
|
60 |
correlation_matrix = df.corr()
|
61 |
-
fig = go.Figure(data=[go.Surface(
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
|
|
|
|
|
|
|
|
|
|
71 |
return fig
|
72 |
|
73 |
-
# Gradio function
|
74 |
def visualize_correlation(selected_series):
|
75 |
# Map descriptive labels back to FRED series IDs
|
76 |
series_ids = [series for series in series_options if series_options[series] in selected_series]
|
@@ -96,7 +108,7 @@ with gr.Blocks() as demo:
|
|
96 |
series_selector = gr.CheckboxGroup(
|
97 |
choices=list(series_options.values()),
|
98 |
label="Select Economic Indicators",
|
99 |
-
info="Choose one or more indicators to include in the correlation matrix."
|
100 |
)
|
101 |
submit_button = gr.Button("Generate Matrix")
|
102 |
|
@@ -104,12 +116,12 @@ with gr.Blocks() as demo:
|
|
104 |
plot_output = gr.Plot(label="3D Correlation Matrix")
|
105 |
error_message = gr.Markdown("", visible=False)
|
106 |
|
107 |
-
# Event handler
|
108 |
submit_button.click(
|
109 |
fn=visualize_correlation,
|
110 |
inputs=[series_selector],
|
111 |
outputs=[plot_output, error_message],
|
112 |
)
|
113 |
|
114 |
-
# Launch the app
|
115 |
demo.launch(debug=True)
|
|
|
7 |
# Set FRED API Key from environment variable
|
8 |
FRED_API_KEY = os.getenv('FRED_API_KEY')
|
9 |
|
10 |
+
# List of FRED data series and their descriptive labels
|
11 |
series_options = {
|
12 |
"UNRATE": "Unemployment Rate",
|
13 |
"GDP": "Gross Domestic Product",
|
|
|
17 |
"M1SL": "M1 Money Supply",
|
18 |
"M2SL": "M2 Money Supply",
|
19 |
"M3SL": "M3 Money Supply",
|
|
|
20 |
"HOUST": "Housing Starts",
|
21 |
"PCE": "Personal Consumption Expenditures",
|
22 |
+
"BAA10YM": "Moody's Baa Corporate Bond Yield Spread"
|
23 |
}
|
24 |
|
25 |
+
# Function to fetch data from the FRED API
|
26 |
def fetch_fred_data(series_ids):
|
27 |
"""
|
28 |
Fetches data for a list of FRED series IDs.
|
|
|
32 |
for series_id in series_ids:
|
33 |
response = requests.get(
|
34 |
f'https://api.stlouisfed.org/fred/series/observations',
|
35 |
+
params={
|
36 |
+
'series_id': series_id,
|
37 |
+
'api_key': FRED_API_KEY,
|
38 |
+
'file_type': 'json'
|
39 |
+
}
|
40 |
)
|
41 |
if response.status_code == 200:
|
42 |
observations = response.json().get('observations', [])
|
43 |
dates = [obs['date'] for obs in observations]
|
44 |
+
# Convert values to float, handling invalid entries
|
45 |
values = [
|
46 |
float(obs['value']) if obs['value'].replace('.', '', 1).isdigit() else float('nan')
|
47 |
for obs in observations
|
|
|
58 |
"""
|
59 |
return (df - df.mean()) / df.std()
|
60 |
|
61 |
+
# Function to create a responsive 3D correlation matrix
|
62 |
def create_3d_correlation_matrix(df):
|
63 |
+
"""
|
64 |
+
Creates a 3D correlation matrix graph using Plotly.
|
65 |
+
The graph will automatically adjust its size.
|
66 |
+
"""
|
67 |
correlation_matrix = df.corr()
|
68 |
+
fig = go.Figure(data=[go.Surface(
|
69 |
+
z=correlation_matrix.values,
|
70 |
+
x=correlation_matrix.columns,
|
71 |
+
y=correlation_matrix.index
|
72 |
+
)])
|
73 |
+
fig.update_layout(
|
74 |
+
title='3D Correlation Matrix (Standardized)',
|
75 |
+
autosize=True, # Enables auto-resizing
|
76 |
+
scene=dict(
|
77 |
+
xaxis=dict(title='Variables'),
|
78 |
+
yaxis=dict(title='Variables'),
|
79 |
+
zaxis=dict(title='Correlation')
|
80 |
+
),
|
81 |
+
margin=dict(l=0, r=0, t=50, b=50) # Adjust margins for better fit
|
82 |
+
)
|
83 |
return fig
|
84 |
|
85 |
+
# Gradio function to handle user interaction
|
86 |
def visualize_correlation(selected_series):
|
87 |
# Map descriptive labels back to FRED series IDs
|
88 |
series_ids = [series for series in series_options if series_options[series] in selected_series]
|
|
|
108 |
series_selector = gr.CheckboxGroup(
|
109 |
choices=list(series_options.values()),
|
110 |
label="Select Economic Indicators",
|
111 |
+
info="Choose one or more indicators to include in the correlation matrix."
|
112 |
)
|
113 |
submit_button = gr.Button("Generate Matrix")
|
114 |
|
|
|
116 |
plot_output = gr.Plot(label="3D Correlation Matrix")
|
117 |
error_message = gr.Markdown("", visible=False)
|
118 |
|
119 |
+
# Event handler for the submit button
|
120 |
submit_button.click(
|
121 |
fn=visualize_correlation,
|
122 |
inputs=[series_selector],
|
123 |
outputs=[plot_output, error_message],
|
124 |
)
|
125 |
|
126 |
+
# Launch the Gradio app
|
127 |
demo.launch(debug=True)
|