relocated functions to utils

app.py

@@ -3,88 +3,14 @@ import torch
 import dash
 import streamlit as st
 import pandas as pd
+import utils
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
 from transformers import pipeline
 from dotenv import load_dotenv
-from plotly.subplots import make_subplots
 import plotly.graph_objects as go
-import plotly.express as px
 
 load_dotenv()
 
-def z_score(y, mean=.04853076, sd=.9409466):
-    return (y - mean) / sd
-
-def indicator_plot(value, title, value_range, domain):
-
-    plot = go.Indicator(
-        mode = "gauge+delta",
-        value = value,
-        domain = domain,
-        title = title,
-        delta = {
-            'reference': 0, 
-            'decreasing': {'color': "#ec4899"},
-            'increasing': {'color': "#36def1"}
-            },
-        gauge = {
-            'axis': {'range': value_range, 'tickwidth': 1, 'tickcolor': "black"},
-            'bar': {'color': "#4361ee"},
-            'bgcolor': "white",
-            'borderwidth': 2,
-            'bordercolor': "#efefef",
-            'steps': [
-                {'range': [value_range[0], 0], 'color': '#efefef'},
-                {'range': [0, value_range[1]], 'color': '#efefef'}
-            ],
-            'threshold': {
-                'line': {'color': "#4361ee", 'width': 8},
-                'thickness': 0.75,
-                'value': value
-            }
-        }
-    )
-
-    return plot
-
-def scatter_plot(df, group_var):
-
-    colors = ['#36def1', '#4361ee'] if group_var else ['#4361ee']
-
-    plot = px.scatter(
-        df, 
-        x='Machine-ratings', 
-        y='Human-ratings',
-        color=group_var,
-        facet_col='x_group', 
-        facet_col_wrap=2,
-        trendline='ols',
-        trendline_scope='trace',
-        hover_data={
-            'Text': df.text,
-            'Language': False,
-            'x_group': False,            
-            'Human-ratings': ':.2f',
-            'Machine-ratings': ':.2f',
-            'Study': df.study,
-            'Instrument': df.instrument,
-        },
-        width=400,
-        height=400,
-        color_discrete_sequence=colors
-    )
-    
-    plot.for_each_annotation(lambda a: a.update(text=a.text.split('=')[-1]))
-    plot.update_layout(
-        legend={
-            'orientation':'h',
-            'yanchor': 'bottom',
-            'y': -.30
-        })
-    plot.update_xaxes(title_standoff = 0)
-
-    return plot
-
 # data import and wrangling
 covariate_columns = {
         'content_domain': 'Content Domain',
@@ -195,16 +121,16 @@ with st.spinner('Processing...'):
 
         with torch.no_grad():
             score = st.session_state.model(**inputs).logits.squeeze().tolist()
-            z = z_score(score)
+            z = utils.z_score(score)
 
-        p1 = indicator_plot(
+        p1 = utils.indicator_plot(
             value=classifier_score,
             title=f'Item Sentiment',
             value_range=[-1, 1],
             domain={'x': [.55, 1], 'y': [0, 1]}        
         )
 
-        p2 = indicator_plot(
+        p2 = utils.indicator_plot(
             value=z,
             title=f'Item Desirability',
             value_range=[-4, 4],
@@ -233,7 +159,6 @@ st.markdown("""
     Figures show the accuarcy in precitions of human-rated item desirability by the sentiment model (left) and the desirability model (right), using `test`-partition data only.
 """)
 
-
 show_covariates = st.checkbox('Show covariates', value=True)
 
 if show_covariates:
@@ -241,6 +166,6 @@ if show_covariates:
 else:
     option = None
 
-plot = scatter_plot(st.session_state.df, option)
+plot = utils.scatter_plot(st.session_state.df, option)
 
 st.plotly_chart(plot, theme=None, use_container_width=True)

utils.py

@@ -0,0 +1,76 @@
+from plotly.subplots import make_subplots
+import plotly.graph_objects as go
+import plotly.express as px
+
+def z_score(y, mean=.04853076, sd=.9409466):
+    return (y - mean) / sd
+
+def indicator_plot(value, title, value_range, domain):
+
+    plot = go.Indicator(
+        mode = "gauge+delta",
+        value = value,
+        domain = domain,
+        title = title,
+        delta = {
+            'reference': 0, 
+            'decreasing': {'color': "#ec4899"},
+            'increasing': {'color': "#36def1"}
+            },
+        gauge = {
+            'axis': {'range': value_range, 'tickwidth': 1, 'tickcolor': "black"},
+            'bar': {'color': "#4361ee"},
+            'bgcolor': "white",
+            'borderwidth': 2,
+            'bordercolor': "#efefef",
+            'steps': [
+                {'range': [value_range[0], 0], 'color': '#efefef'},
+                {'range': [0, value_range[1]], 'color': '#efefef'}
+            ],
+            'threshold': {
+                'line': {'color': "#4361ee", 'width': 8},
+                'thickness': 0.75,
+                'value': value
+            }
+        }
+    )
+
+    return plot
+
+def scatter_plot(df, group_var):
+
+    colors = ['#36def1', '#4361ee'] if group_var else ['#4361ee']
+
+    plot = px.scatter(
+        df, 
+        x='Machine-ratings', 
+        y='Human-ratings',
+        color=group_var,
+        facet_col='x_group', 
+        facet_col_wrap=2,
+        trendline='ols',
+        trendline_scope='trace',
+        hover_data={
+            'Text': df.text,
+            'Language': False,
+            'x_group': False,            
+            'Human-ratings': ':.2f',
+            'Machine-ratings': ':.2f',
+            'Study': df.study,
+            'Instrument': df.instrument,
+        },
+        width=400,
+        height=400,
+        color_discrete_sequence=colors
+    )
+    
+    plot.for_each_annotation(lambda a: a.update(text=a.text.split('=')[-1]))
+    plot.update_layout(
+        legend={
+            'orientation':'h',
+            'yanchor': 'bottom',
+            'y': -.30
+        })
+    plot.update_xaxes(title_standoff = 0)
+
+    return plot