Update app.py

app.py

@@ -1,75 +1,41 @@
-import streamlit as st
+import gradio as gr
+import tensorflow as tf
+from huggingface_hub import from_pretrained_keras
 import numpy as np
-import matplotlib.pyplot as plt
-from sklearn.linear_model import LinearRegression, Ridge
-from sklearn.preprocessing import PolynomialFeatures
-from sklearn.metrics import mean_squared_error
 
-st.subheader("Ridge Demo")
-col1, col2 = st.columns(2)
+model = from_pretrained_keras("keras-io/semi-supervised-classification-simclr")
 
-degree = st.slider('Degree', 2, 40, 1)
-alpha = st.slider('Lambda (Regularisation)', 0, 500, 1)
+labels = ["airplane", "bird", "car", "cat", "deer", "dog", "horse", "monkey", "ship", "truck"]
 
 
-with col1:
-    st.markdown("#### Un-regularized")
-    
-with col2:
-    st.markdown("#### Regularized")
-    
-x = np.linspace(-1., 1., 100)
-y = 4 + 3*x + 2*np.sin(x) + 2*np.random.randn(len(x))
+def infer(test_image):
+    image = tf.constant(test_image)
+    image = tf.reshape(image, [-1, 96, 96, 3])
+    pred = model.predict(image)
+    pred_list = pred[0, :]
+    pred_softmax = np.exp(pred_list)/np.sum(np.exp(pred_list))
+    softmax_list = pred_softmax.tolist()
+    return {labels[i]: softmax_list[i] for i in range(10)}
 
 
-poly = PolynomialFeatures(degree=degree, include_bias=False)
-x_new = poly.fit_transform(x.reshape(-1, 1))
+image = gr.inputs.Image(shape=(96, 96))
+label = gr.outputs.Label(num_top_classes=3)
 
-lr = LinearRegression()
-lr.fit(x_new, y)
-y_pred = lr.predict(x_new)
 
+article = """<center>
+Authors: <a href='https://twitter.com/johko990' target='_blank'>Johannes Kolbe</a> after an example by András Béres at 
+<a href='https://keras.io/examples/vision/semisupervised_simclr/' target='_blank'>keras.io</a>"""
 
-ri = Ridge(alpha = alpha)
-ri.fit(x_new, y)
-y_pred_ri = ri.predict(x_new)
 
+description = """Image classification with a model trained via Semi-supervised Contrastive Learning """
 
-fig1, ax1 = plt.subplots()
-fig2, ax2 = plt.subplots()
 
-ax1.scatter(x, y)
-ax1.plot(x, y_pred)
-
-ax2.scatter(x, y)
-ax2.plot(x, y_pred_ri)
-
-for ax in [ax1, ax2]:
-    ax.spines['right'].set_visible(False)
-    ax.spines['top'].set_visible(False)
-
-    # Only show ticks on the left and bottom spines
-    ax.yaxis.set_ticks_position('left')
-    ax.xaxis.set_ticks_position('bottom')
-
-    ax.set_xlabel("x")
-    ax.set_ylabel("y")
-
-rmse = np.round(np.sqrt(mean_squared_error(y_pred, y)), 2)
-ax1.set_title(f"Train RMSE: {rmse}")
-
-rmse_ri = np.round(np.sqrt(mean_squared_error(y_pred_ri, y)), 2)
-ax2.set_title(f"Train RMSE: {rmse_ri}")
-
-with col1:
-    st.pyplot(fig1)
-    
-with col2:
-    st.pyplot(fig2)
-hide_streamlit_style = """
-            <style>
-            #MainMenu {visibility: hidden;}
-            footer {visibility: hidden;}
-            </style>
-            """
-st.markdown(hide_streamlit_style, unsafe_allow_html=True) 
+Iface = gr.Interface(
+    fn=infer,
+    inputs=image,
+    outputs=label,
+    examples=[["examples/monkey.jpeg"], ["examples/titanic.jpg"], ["examples/truck.jpg"]],
+    title="Semi-Supervised Contrastive Learning Classification",
+    article=article,
+    description=description,
+).launch()