Added first working version of application, NAPSI is dummy.

.gitignore

@@ -0,0 +1,2 @@
+.vscode/launch.json
+*.pyc

DummyModel.py

@@ -7,7 +7,7 @@ class DummyModel(torch.nn.Module):
         super().__init__()
 
     def forward(self, x):
-        return 1
+        return torch.softmax(torch.rand(5), 0)
 
     def __call__(self, x):
         return self.forward(x)

app.py

@@ -1,26 +1,36 @@
-import torch
-import os
-
 import gradio as gr
-from huggingface_hub import hf_hub_download
 
-from DummyModel import DummyModel
+from backend import Infer
+
+
+DEBUG = True
 
 
-file_path = hf_hub_download("lfolle/DeepNAPSIModel", "dummy_model.pth",
-                            use_auth_token=os.environ['DeepNAPSIModel'])
-model = DummyModel()
-model.load_state_dict(torch.load(file_path))
+infer = Infer(DEBUG)
 
-def predict(data):
-    return model(data)
+with gr.Blocks(analytics_enabled=False, title="DeepNAPSI Prediction") as demo:
+    with gr.Column():
+        gr.Markdown("Upload an image of the hand and click **Predict NAPSI** to see the output.")
+        with gr.Column():
+            image_input = gr.Image()
+            image_button = gr.Button("Predict NAPSI")
+            outputs = []
+            with gr.Row():
+                with gr.Column():
+                    outputs.append(gr.Image())
+                    outputs.append(gr.Number(label="DeepNAPSI Thumb"))
+                with gr.Column():
+                    outputs.append(gr.Image())
+                    outputs.append(gr.Number(label="DeepNAPSI Index"))
+                with gr.Column():
+                    outputs.append(gr.Image())
+                    outputs.append(gr.Number(label="DeepNAPSI Middle"))
+                with gr.Column():
+                    outputs.append(gr.Image())
+                    outputs.append(gr.Number(label="DeepNAPSI Ring"))
+                with gr.Column():
+                    outputs.append(gr.Image())
+                    outputs.append(gr.Number(label="DeepNAPSI Pinky"))
+        image_button.click(infer.predict, inputs=image_input, outputs=outputs)
 
-iface = gr.Interface(
-    predict,
-    title="DeepNAPSI Application",
-    inputs=gr.Image(),
-    outputs=gr.Number(label="DeepNAPSI prediction"),
-    description="",
-    examples=["assets/hand_example.jpg"],
-)
-iface.launch()
+demo.launch(share=True)

backend.py

@@ -0,0 +1,34 @@
+import torch
+import cv2
+import numpy as np
+from huggingface_hub import hf_hub_download
+from nail_detection.main import get_nails
+
+from DummyModel import DummyModel
+
+
+def load_model(DEBUG):
+    model = DummyModel()
+    if not DEBUG:
+        file_path = hf_hub_download("lfolle/DeepNAPSIModel", "dummy_model.pth",
+                                    use_auth_token=os.environ['DeepNAPSIModel'])
+        model.load_state_dict(torch.load(file_path))
+    return model
+
+
+class Infer():
+    def __init__(self, DEBUG):
+        self.model = load_model(DEBUG)
+
+    def predict(self, data):
+        nails = get_nails(cv2.cvtColor(data, cv2.COLOR_RGB2BGR))
+        predictions = []
+        if nails is None:
+            for _ in range(5):
+                predictions.append(np.zeros((64, 64, 3)))
+                predictions.append(-1)
+        else:
+            for nail in nails:
+                predictions.append(nail)
+                predictions.append(int(torch.argmax(self.model(nail))))
+        return predictions

requirements.txt

@@ -1,2 +1,5 @@
-torch
-numpy
+cvs2
+numpy
+gradio
+huggingface_hub
+torch