Synced repo using 'sync_with_huggingface' Github Action

app.py

@@ -0,0 +1,46 @@
+import pandas as pd
+import streamlit as st
+
+
+URL_BASE = "https://drivendata-competition-meta-vsc-data-us.s3.us-west-2.amazonaws.com"
+URL_GROUND_TRUTH_FILE = f"{URL_BASE}/train/train_descriptor_ground_truth.csv"
+URL_QUERY_BASE = f"{URL_BASE}/train/query/"
+URL_REF_BASE = f"{URL_BASE}/train/reference/"
+
+
+@st.cache
+def load_data():
+    df = pd.read_csv(URL_GROUND_TRUTH_FILE)
+    query_ids = df.query_id.unique().tolist()
+    ref_ids = df.ref_id.unique().tolist()
+
+    query_to_refs = {x: [] for x in query_ids}
+    ref_to_queries = {x: [] for x in ref_ids}
+    for _, row in df.iterrows():
+        query_to_refs[row.query_id].append(row.ref_id)
+        ref_to_queries[row.ref_id].append(row.query_id)
+    return query_to_refs, ref_to_queries
+
+
+st.title("Video Similarity Challenge Ground Truth Visualizer")
+
+q_to_r, r_to_q = load_data()
+
+with st.sidebar:
+    search_by = st.selectbox("Search By:", ["query", "reference"])
+    chosen_val = st.selectbox("Select ID:", q_to_r.keys() if search_by == "query" else r_to_q.keys())
+
+results_by = "reference" if search_by == "query" else "query"
+search_url = URL_QUERY_BASE if search_by == "query" else URL_REF_BASE
+results_url = URL_REF_BASE if search_by == "query" else URL_QUERY_BASE
+
+col1, col2 = st.columns(2)
+
+with col1:
+    st.write(f"{search_by.title()} ID: {chosen_val}")
+    st.video(f"{search_url}{chosen_val}.mp4")
+
+with col2:
+    for i, vid in enumerate(q_to_r[chosen_val] if search_by == "query" else r_to_q[chosen_val]):
+        st.write(f"{results_by.title()} ID: {vid}")
+        st.video(f"{results_url}{vid}.mp4")

requirements.txt

@@ -0,0 +1,2 @@
+streamlit
+pandas