build(docker): switch to docker sdk

Dockerfile

@@ -0,0 +1,49 @@
+# Builder stage
+FROM python:3.11-alpine AS builder
+
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+# Enable bytecode compilation
+ENV UV_COMPILE_BYTECODE=1
+# Copy from the cache instead of linking since it's a mounted volume
+ENV UV_LINK_MODE=copy
+# Install to a specific directory that we can copy later
+# cf. https://github.com/astral-sh/uv/issues/8085#issuecomment-2438256688
+ENV UV_PROJECT_ENVIRONMENT="/opt/venv"
+
+RUN apk add --no-cache gcc python3-dev musl-dev linux-headers git
+# Install the project's dependencies using the lockfile and settings
+# https://docs.astral.sh/uv/guides/integration/docker/#using-uv-temporarily
+RUN --mount=from=ghcr.io/astral-sh/uv:0.9.5,source=/uv,target=/bin/uv \
+    --mount=type=cache,target=/root/.cache/uv \
+    --mount=type=bind,source=uv.lock,target=uv.lock \
+    --mount=type=bind,source=pyproject.toml,target=pyproject.toml \
+    uv pip install -r requirements.txt
+
+# Runtime stage
+FROM python:3.11-alpine AS runtime
+
+WORKDIR /app
+
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONPATH="/app"
+# Set the path to use our installed packages
+ENV PATH="/opt/venv/bin:$PATH"
+ENV PYTHON_PATH="/opt/venv/lib/python3.11/site-packages"
+
+LABEL maintainer="F-G Fernandez <[email protected]>"
+
+# Install only curl for healthcheck
+RUN apk add --no-cache curl
+
+# Copy installed dependencies from builder
+COPY --from=builder /opt/venv /opt/venv
+
+# Copy project code
+COPY src ./src
+
+# Entrypoint
+EXPOSE 8501
+HEALTHCHECK --interval=10s --timeout=3s --retries=5 CMD ["curl", "-f", "http://localhost:8501/_stcore/health", "--max-time", "3"]
+ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

README.md

@@ -3,8 +3,10 @@ title: TorchCAM
 emoji: 🎨
 colorFrom: purple
 colorTo: pink
-sdk: streamlit
-app_file: app.py
+sdk: docker
+app_port: 8501
+tags:
+  - streamlit
 pinned: true
 license: apache-2.0
 thumbnail: >-

app.py → src/streamlit_app.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021-2023, François-Guillaume Fernandez.
+# Copyright (C) 2021-2025, François-Guillaume Fernandez.
 
 # This program is licensed under the Apache License 2.0.
 # See LICENSE or go to <https://www.apache.org/licenses/LICENSE-2.0> for full license details.
@@ -17,7 +17,17 @@ from torchcam import methods
 from torchcam.methods._utils import locate_candidate_layer
 from torchcam.utils import overlay_mask
 
-CAM_METHODS = ["CAM", "GradCAM", "GradCAMpp", "SmoothGradCAMpp", "ScoreCAM", "SSCAM", "ISCAM", "XGradCAM", "LayerCAM"]
+CAM_METHODS = [
+    "CAM",
+    "GradCAM",
+    "GradCAMpp",
+    "SmoothGradCAMpp",
+    "ScoreCAM",
+    "SSCAM",
+    "ISCAM",
+    "XGradCAM",
+    "LayerCAM",
+]
 TV_MODELS = [
     "resnet18",
     "resnet50",
@@ -28,14 +38,14 @@ TV_MODELS = [
     "convnext_small",
 ]
 LABEL_MAP = requests.get(
-    "https://raw.githubusercontent.com/anishathalye/imagenet-simple-labels/master/imagenet-simple-labels.json"
+    "https://raw.githubusercontent.com/anishathalye/imagenet-simple-labels/master/imagenet-simple-labels.json",
+    timeout=10,
 ).json()
 
 
 def main():
-
     # Wide mode
-    st.set_page_config(layout="wide")
+    st.set_page_config(page_title="TorchCAM - Class activation explorer", layout="wide")
 
     # Designing the interface
     st.title("TorchCAM: class activation explorer")
@@ -50,14 +60,12 @@ def main():
     # Sidebar
     # File selection
     st.sidebar.title("Input selection")
-    # Disabling warning
-    st.set_option("deprecation.showfileUploaderEncoding", False)
     # Choose your own image
     uploaded_file = st.sidebar.file_uploader("Upload files", type=["png", "jpeg", "jpg"])
     if uploaded_file is not None:
         img = Image.open(BytesIO(uploaded_file.read()), mode="r").convert("RGB")
 
-        cols[0].image(img, use_column_width=True)
+        cols[0].image(img, use_container_width=True)
 
     # Model selection
     st.sidebar.title("Setup")
@@ -87,25 +95,28 @@ def main():
     )
     if cam_method is not None:
         cam_extractor = methods.__dict__[cam_method](
-            model, target_layer=[s.strip() for s in target_layer.split("+")] if len(target_layer) > 0 else None
+            model,
+            target_layer=[s.strip() for s in target_layer.split("+")] if len(target_layer) > 0 else None,
         )
 
     class_choices = [f"{idx + 1} - {class_name}" for idx, class_name in enumerate(LABEL_MAP)]
-    class_selection = st.sidebar.selectbox("Class selection", ["Predicted class (argmax)"] + class_choices)
+    class_selection = st.sidebar.selectbox("Class selection", ["Predicted class (argmax)", *class_choices])
 
     # For newline
     st.sidebar.write("\n")
 
     if st.sidebar.button("Compute CAM"):
-
         if uploaded_file is None:
             st.sidebar.error("Please upload an image first")
 
         else:
             with st.spinner("Analyzing..."):
-
                 # Preprocess image
-                img_tensor = normalize(to_tensor(resize(img, (224, 224))), [0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+                img_tensor = normalize(
+                    to_tensor(resize(img, (224, 224))),
+                    [0.485, 0.456, 0.406],
+                    [0.229, 0.224, 0.225],
+                )
 
                 if torch.cuda.is_available():
                     img_tensor = img_tensor.cuda()