Switch to gradio app

Dockerfile

@@ -7,23 +7,27 @@ RUN apt-get install -y sudo wget curl nano git \
 # Create a group and user
 ENV APPUSER="appuser"
 ENV HOME=/home/$APPUSER
-RUN groupadd -g 999 appgroup && \
-    useradd -r -u 999 -g appgroup $APPUSER
+RUN useradd -m -u 1000 $APPUSER
 
 USER $APPUSER
 WORKDIR $HOME
 
-RUN pip install --user streamlit==1.28
+# Set home to the user's home directory
+ENV HOME=/home/$APPUSER \
+    PATH=/home/$APPUSER/.local/bin:$PATH
 
-ENV PATH="$HOME/.local/bin:$PATH"
+COPY --chown=$APPUSER ./requirements.txt $HOME/app/requirements.txt
+WORKDIR $HOME/app
 
-COPY --chown=$APPUSER:appgroup . $HOME/app
+RUN pip install --no-cache-dir --user -r requirements.txt
 
-WORKDIR $HOME/app
+COPY --chown=$APPUSER . $HOME/app
 
-# Expose port for streamlit
-ENV PORT=8501
+# Expose port for web service
+ENV PORT=7860
 EXPOSE $PORT
 
 # Run streamlit app under conda environment
-CMD ["sh", "-c", "streamlit run --server.port=$PORT --server.address=0.0.0.0 app.py"]
+# CMD ["sh", "-c", "streamlit run --server.port=$PORT --server.address=0.0.0.0 app.py"]
+
+CMD ["python", "main.py"]

main.py

@@ -0,0 +1,30 @@
+import gradio as gr
+import torch
+import requests
+from torchvision import transforms
+
+model = torch.hub.load("pytorch/vision:v0.6.0", "resnet18", pretrained=True).eval()
+response = requests.get("https://git.io/JJkYN")
+labels = response.text.split("\n")
+
+
+def predict(inp):
+    inp = transforms.ToTensor()(inp).unsqueeze(0)
+    with torch.no_grad():
+        prediction = torch.nn.functional.softmax(model(inp)[0], dim=0)
+        confidences = {labels[i]: float(prediction[i]) for i in range(1000)}
+    return confidences
+
+
+def run():
+    demo = gr.Interface(
+        fn=predict,
+        inputs=gr.inputs.Image(type="pil"),
+        outputs=gr.outputs.Label(num_top_classes=3),
+    )
+
+    demo.launch(server_name="0.0.0.0", server_port=7860)
+
+
+if __name__ == "__main__":
+    run()

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+torch
+torchvision
+requests