updates demos

demo/fraud_detection.py

@@ -1,37 +0,0 @@
-import gradio as gr
-
-def fraud_detection(income, assets):
-    income *= 1000
-    total_asset_cost = sum(assets["Cost"])
-    total_deductible = sum(assets[assets["Deduct"]]["Cost"])
-    if total_asset_cost > income / 2:
-        fraud_prob = max(total_asset_cost / income, total_deductible / total_asset_cost)
-    else: 
-        # Benford's law
-        expected_digit_distribution = {1: .301, 2: .176, 3: .125}
-        numbers = [income] + list(assets["Cost"])
-        digit_distribution = {i: 0 for i in range(10)}
-        for number in numbers:
-            leading_digit = str(int(number))[0]
-            digit_distribution[int(leading_digit)] += 1
-        fraud_prob = 0
-        for digit, expected in expected_digit_distribution.items():
-            fraud_prob += abs(digit_distribution[digit] / len(numbers) - expected) / 10
-
-    fraud_prob = min(1, fraud_prob)
-    return {"fraud": fraud_prob, "standard": 1 - fraud_prob}
-
-io = gr.Interface(
-    fraud_detection, 
-    [
-        gr.inputs.Slider(1, 500, label="Income (in thousands)"),
-        gr.inputs.Dataframe(
-            headers=["Item", "Cost", "Deduct"], 
-            datatype=["str", "number", "bool"],
-            label="Assets Purchased this Year"
-        )
-    ],
-    "label"
-)
-
-io.launch()

demo/image_classifier.py

@@ -1,7 +1,7 @@
+# Demo: (Image) -> (Label)
+
 import gradio as gr
 import tensorflow as tf
-# from vis.utils import utils
-# from vis.visualization import visualize_cam
 
 import numpy as np
 from PIL import Image

demo/log.log

@@ -1 +0,0 @@
-++++++++++++

demo/longest_word.py

@@ -1,8 +1,7 @@
+# Demo: (Textbox) -> (Label)
+
 import gradio as gr
 
-gr.reset_all()
-
-
 def longest_word(text):
     words = text.split(" ")
     lengths = [len(word) for word in words]

demo/sentiment_analysis.py

@@ -1,3 +1,5 @@
+# Demo: (Textbox) -> (Label)
+
 import gradio as gr
 import nltk
 from nltk.sentiment.vader import SentimentIntensityAnalyzer

demo/tax_calculator.py

@@ -0,0 +1,43 @@
+# Demo: (Number, Radio, Dataframe) -> (Textbox)
+
+import gradio as gr
+
+def tax_calculator(income, marital_status, assets):
+    tax_brackets = [
+        (10, 0),
+        (25, 8),
+        (60, 12),
+        (120, 20),
+        (250, 30)
+    ]
+    total_deductible = sum(assets[assets["Deduct"]]["Cost"])
+    taxable_income = income - total_deductible
+
+    total_tax = 0
+    for bracket, rate in tax_brackets:
+        if taxable_income > bracket:
+            total_tax += (taxable_income - bracket) * rate / 100
+    
+    if marital_status == "Married":
+        total_tax *= 0.75
+    elif marital_status == "Divorced":
+        total_tax *= 0.8
+    
+    return round(total_tax)
+
+io = gr.Interface(
+    tax_calculator, 
+    [
+        "number",
+        gr.inputs.Radio(["Single", "Married", "Divorced"]),
+        gr.inputs.Dataframe(
+            headers=["Item", "Cost", "Deduct"], 
+            datatype=["str", "number", "bool"],
+            label="Assets Purchased this Year"
+        )
+    ],
+    "number",
+    interpretation="default"
+)
+
+io.launch()

demo/titanic_survival.py

@@ -1,3 +1,5 @@
+# Demo: (Dropdown, Checkbox, Slider, CheckboxGroup, Number, Radio) -> (Label)
+
 import pandas as pd
 import numpy as np
 import sklearn
@@ -71,7 +73,7 @@ io = gr.Interface(
         "checkbox",
         gr.inputs.Slider(0, 80),
         gr.inputs.CheckboxGroup(["Sibling", "Child"], label="Travelling with (select all)"),
-        gr.inputs.Slider(0, 150),
+        gr.inputs.Number().interpret(delta_type="absolute", delta=5),
         gr.inputs.Radio(["S", "C", "Q"], type="index"),
     ],
     "label",