Upload folder using huggingface_hub

.dockerignore

@@ -0,0 +1,26 @@
+# Ignore Python cache files
+__pycache__/
+*.pyc
+*.pyo
+
+# Ignore virtual environments
+venv/
+env/
+.venv/
+
+# Ignore version control system directories
+.git/
+.gitignore
+
+# Ignore Docker-related files
+Dockerfile
+.dockerignore
+
+# Ignore IDE/editor-specific files
+.vscode/
+.idea/
+
+# Ignore any additional temporary files
+*.tmp
+*.log
+*.swp

Dockerfile

@@ -0,0 +1,20 @@
+# Use the official Python image from the Docker Hub
+FROM python:3.8-slim
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy the requirements.txt file into the container
+COPY requirements.txt .
+
+# Install the dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the rest of the application code into the container
+COPY . .
+
+# Expose the port the app runs on
+EXPOSE 8501
+
+# Command to run the Streamlit app
+CMD ["streamlit", "run", "streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

app.py

@@ -0,0 +1,42 @@
+import requests
+import streamlit as st
+from streamlit_lottie import st_lottie
+from bokeh.embed import components
+from bokeh_plot import create_plot
+
+@st.cache_data()
+def load_lottieurl(url: str):
+    r = requests.get(url)
+    if r.status_code != 200:
+        return None
+    return r.json()
+
+st.set_page_config(
+    page_title="AToMiC2024 Images (Sampled 50k)", 
+    page_icon="⚛️", 
+    layout="wide", 
+    initial_sidebar_state="auto", 
+    menu_items={'About': '## UMAP Embeddings  of AToMiC2024 images'}
+)
+
+if __name__ == "__main__":
+    col1, col2 = st.columns([0.15, 0.85])
+    with col1:
+        lottie = load_lottieurl("https://lottie.host/de47fd4c-99cb-48a7-ae10-59d4eb8e4dbe/bXMpZN95tA.json")
+        st_lottie(lottie)
+
+    with col2:
+        st.write(
+            """
+            ## AToMiC Image Explorer
+            ### Subsampled AToMiC Images using [CLIP-ViT-BigG](https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k)
+            - **Subsampling Procedure:** Hierarchical K-Means [10, 10, 10, 10], randomly sampled 50 from the leaf clusters -> random sample 25k for visualization.
+            - Original [Image Collection](https://huggingface.co/datasets/TREC-AToMiC/AToMiC-Images-v0.2)
+            - Prebuilt [Embeddings/Index](https://huggingface.co/datasets/TREC-AToMiC/AToMiC-Baselines/tree/main/indexes)
+            - Questions? Leave an issue at our [repo](https://github.com/TREC-AToMiC/AToMiC).
+            - It takes a few minutes to render the plot.
+            """
+        )
+    # Generate the Bokeh plot
+    bokeh_plot = create_plot()
+    st.bokeh_chart(bokeh_plot, use_container_width=False)

bokeh_plot.py

@@ -0,0 +1,59 @@
+import pickle
+import pandas as pd
+from bokeh.plotting import figure
+from bokeh.models import ColumnDataSource, HoverTool, Div
+from bokeh.layouts import column
+import logging
+
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+def create_plot():
+    # Load preprocessed data
+    logger.info("Loading preprocessed data...")
+    with open('processed_data.pkl', 'rb') as f:
+        df = pickle.load(f)
+    with open('contour_data.pkl', 'rb') as f:
+        contour_data = pickle.load(f)
+    logger.info("Data loaded successfully.")
+    
+    logger.info("Creating Bokeh plot...")
+    
+    p = figure(width=1280, height=800, title="UMAP projection of embeddings")
+
+    # Load contour data
+    contour_source = ColumnDataSource(data=dict(xs=contour_data['xs'], 
+                                                ys=contour_data['ys'], 
+                                                color=contour_data['color']))
+    contour_renderer = p.patches(xs="xs", ys="ys", source=contour_source, fill_alpha=0.3, line_color=None, fill_color="color")
+
+    # Scatter plot
+    source = ColumnDataSource(df)
+    scatter_renderer = p.scatter('x', 'y', size=3, source=source, fill_alpha=0.2, line_alpha=0.1)
+
+    # Configure hover tool to display images
+    hover = HoverTool(renderers=[scatter_renderer])
+    hover.tooltips = """
+        <div>
+            <div>
+                <strong>Image ID:</strong> @id
+            </div>
+            <div>
+                <strong>Cap_ref:</strong> @caption
+            </div>
+            <div>
+                <strong>URL:</strong> @url
+            </div>
+            <div>
+                <img
+                    src="data:image/jpeg;base64,@image_b64" height="200" alt="Image"
+                    style="float: left; margin: 0px 15px 15px 0px;"
+                    border="2"
+                ></img>
+            </div>
+        </div>
+    """
+    p.add_tools(hover)
+
+    return p

config.toml

@@ -0,0 +1,3 @@
+[server]
+maxUploadSize = 1024
+maxMessageSize = 500

contour_data.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a992bb94f2a6d982d266b4e9cac4ae601bf1dcb5a0d99c1fa5d02da3d069bfc6
+size 386638

processed_data.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:364112f46defd36ce1c8d3e4e49e48ba2e27e4863ffd65376814ab28611d4fc0
+size 94337630

requirements.txt

@@ -0,0 +1,5 @@
+bokeh==2.4.3
+pandas
+numpy
+streamlit
+streamlit-lottie