new app

app.py

@@ -1,118 +1,114 @@
 import gradio as gr
-#example just for fun 
-from src.visual_qa import main_text_retrieve_images
-from src.generate_queries_alternative import main_generate_queries
-import time
 import pandas as pd
-
-import spacy
-
-# Try to load the model, and download it if missing
-try:
-    nlp = spacy.load("en_core_web_sm")
-except OSError:
-    from spacy.cli import download
-    download("en_core_web_sm")
-    nlp = spacy.load("en_core_web_sm")
-
-
-
-def handle_structured_query(question, sort_by=""):
-    if not question:
-        return "Please ask something 🙂", pd.DataFrame(), []
-
-    try:
-        start = time.time()
-        result_query, sparql_query = main_generate_queries(question)
-        elapsed = round(time.time() - start, 2)
-    except Exception as e:
-        return f"⚠️ Query failed: {e}", pd.DataFrame(), []
-
-    if isinstance(result_query, str):
-        return result_query, pd.DataFrame(), []
-
-    if not result_query:
-        return f"No results for '{question}'. Try rephrasing. (⏱ {elapsed}s)", pd.DataFrame(), []
-
-    df = pd.DataFrame(result_query)
-    if sort_by and sort_by in df.columns:
-        df = df.sort_values(by=sort_by)
-
-
-    if "image_url" in df.columns:
-        columns_of_interest = ["image_url", "year","fashion_collectionLabel", "reference_URL"]
-        df = df[columns_of_interest]
-        # Create a gallery: each item is (image_url, metadata string)
-        gallery_items = []
-        for _, row in df.iterrows():
-            image_url = row.get("image_url")
-            if not image_url:
-                continue
-            # Caption from other fields
-            caption = " | ".join(f"{k}: {v}" for k, v in row.items() if k != "image_url" and pd.notnull(v))
-            gallery_items.append((image_url, caption))
-        return f"Query returned {len(gallery_items)} image(s) in {elapsed} seconds.", pd.DataFrame(), gallery_items
-
-    return f"Query returned a table with {len(df)} row(s) in {elapsed} seconds.", df, []
-
-
-
-
-def handle_image_query(text):
-    if not text:
-        return []
-
-    try:
-        records = main_text_retrieve_images(text)
-        print(f"Retrieved {len(records)} records for query: {text}")
-        print(records)
-    except Exception as e:
-        return [("https://via.placeholder.com/300x200?text=Error", f"Error: {e}")]
-
-    gallery_items = []
-    for item in records:
-        image_url = item.get("image_url")
-        if not image_url:
-            continue
-        # Build a simple caption from the remaining fields
-        caption = " | ".join(f"{k}: {v}" for k, v in item.items() if k != "image_url")
-        gallery_items.append((image_url, caption))
-
-    return gallery_items
-
-# --- UI --- #
+import numpy as np
+from search import search_images_by_text, get_similar_images
+import requests
+from io import BytesIO
+
+import requests
+from io import BytesIO
+#@st.cache_data(show_spinner="Loading FashionDB...")
+def load_data_hf():
+    # Load the Parquet file directly from Hugging Face
+    df_url = "https://huggingface.co/datasets/traopia/vogue_runway_small/resolve/main/VogueRunway.parquet"
+    df = pd.read_parquet(df_url)
+
+    # Load the .npy file using requests
+    npy_url = "https://huggingface.co/datasets/traopia/vogue_runway_small/resolve/main/VogueRunway_image.npy"
+    response = requests.get(npy_url)
+    response.raise_for_status()  # Raise error if download fails
+    embeddings = np.load(BytesIO(response.content),  mmap_mode="r")
+
+    return df, embeddings
+
+df, embeddings = load_data_hf()
+
+# Filter and search
+def filter_and_search(fashion_house, category, season, year_range, query):
+    filtered = df.copy()
+
+    if fashion_house:
+        filtered = filtered[filtered['designer'].isin(fashion_house)]
+    if category:
+        filtered = filtered[filtered['category'].isin(category)]
+    if season:
+        filtered = filtered[filtered['season'].isin(season)]
+    filtered = filtered[(filtered['year'] >= year_range[0]) & (filtered['year'] <= year_range[1])]
+
+    if query:
+        results = search_images_by_text(query, filtered, embeddings)
+    else:
+        results = filtered.head(30)
+    
+    image_urls = results["url"].tolist()
+    metadata = results.to_dict(orient="records")
+    return image_urls, metadata
+
+# Display metadata and similar
+def show_metadata(idx, metadata):
+    item = metadata[idx]
+    out = ""
+    for field in ["designer", "season", "year", "category"]:
+        if field in item and pd.notna(item[field]):
+            out += f"**{field.title()}**: {item[field]}\n"
+    if 'collection' in item and pd.notna(item['collection']):
+        out += f"\n[View Collection]({item['collection']})"
+    return out
+
+def find_similar(idx, metadata):
+    key = metadata[idx]["key"]
+    similar_df = get_similar_images(df, key, embeddings, top_k=5)
+    return similar_df["url"].tolist(), similar_df.to_dict(orient="records")
+
+# Gradio UI
 with gr.Blocks() as demo:
-    gr.Markdown("# 🧵 FashionDB Interface")
-
-
-    with gr.Tab("Structured Query"):
-        gr.Markdown("Ask FashionDB anything and view results with images + metadata.")
-
-        with gr.Row():
-            query_input = gr.Textbox(label="Your question")
-            sort_input = gr.Textbox(label="Sort by (optional column name)", placeholder="e.g. start_year")
-
-        query_submit = gr.Button("Submit")
-
-        query_text_output = gr.Textbox(label="Message", interactive=False)
-        query_table_output = gr.Dataframe(label="Tabular Result", interactive=False)
-        query_gallery_output = gr.Gallery(label="Image Gallery")
-        query_submit.click(
-            fn=handle_structured_query,
-            inputs=[query_input, sort_input],
-            outputs=[
-                query_text_output,
-                query_table_output,
-                query_gallery_output
-            ]
-        )
-
-    with gr.Tab("Image Retrieval"):
-        gr.Markdown("Search for similar fashion show images based on a text description.")
-        image_text = gr.Textbox(label="Describe the kind of images you're looking for")
-        image_submit = gr.Button("Find Images")
-        image_gallery = gr.Gallery(label="Retrieved Images")
-
-        image_submit.click(handle_image_query, inputs=image_text, outputs=image_gallery)
+    gr.Markdown("# 👗 FashionDB Explorer")
+
+    with gr.Row():
+        fashion_house = gr.Dropdown(label="Fashion House", choices=sorted(df["designer"].dropna().unique()), multiselect=True)
+        category = gr.Dropdown(label="Category", choices=sorted(df["category"].dropna().unique()), multiselect=True)
+        season = gr.Dropdown(label="Season", choices=sorted(df["season"].dropna().unique()), multiselect=True)
+        year_range = gr.Slider(label="Year Range", minimum=int(df['year'].min()), maximum=int(df['year'].max()), value=(2000, 2025), step=1)
+
+    query = gr.Textbox(label="Search", placeholder="e.g., pink dress")
+    search_button = gr.Button("Search")
+    
+    result_gallery = gr.Gallery(label="Search Results").style(grid=[5], height="auto")
+    metadata_output = gr.Markdown()
+    similar_gallery = gr.Gallery(label="Similar Images").style(grid=[5], height="auto")
+
+    metadata_state = gr.State([])
+    selected_idx = gr.Number(value=0, visible=False)
+
+    def handle_search(*args):
+        imgs, meta = filter_and_search(*args)
+        return imgs, meta, "", []
+
+    search_button.click(
+        handle_search,
+        inputs=[fashion_house, category, season, year_range, query],
+        outputs=[result_gallery, metadata_state, metadata_output, similar_gallery]
+    )
+
+    def handle_click(evt: gr.SelectData, metadata):
+        idx = evt.index
+        md = show_metadata(idx, metadata)
+        return idx, md
+
+    result_gallery.select(
+        handle_click,
+        inputs=[metadata_state],
+        outputs=[selected_idx, metadata_output]
+    )
+
+    def show_similar(idx, metadata):
+        return find_similar(int(idx), metadata)
+
+    show_similar_button = gr.Button("Show Similar Images")
+    show_similar_button.click(
+        show_similar,
+        inputs=[selected_idx, metadata_state],
+        outputs=[similar_gallery, metadata_state]
+    )
 
 demo.launch()

app_old.py

@@ -0,0 +1,118 @@
+import gradio as gr
+#example just for fun 
+from src.visual_qa import main_text_retrieve_images
+from src.generate_queries_alternative import main_generate_queries
+import time
+import pandas as pd
+
+import spacy
+
+# Try to load the model, and download it if missing
+try:
+    nlp = spacy.load("en_core_web_sm")
+except OSError:
+    from spacy.cli import download
+    download("en_core_web_sm")
+    nlp = spacy.load("en_core_web_sm")
+
+
+
+def handle_structured_query(question, sort_by=""):
+    if not question:
+        return "Please ask something 🙂", pd.DataFrame(), []
+
+    try:
+        start = time.time()
+        result_query, sparql_query = main_generate_queries(question)
+        elapsed = round(time.time() - start, 2)
+    except Exception as e:
+        return f"⚠️ Query failed: {e}", pd.DataFrame(), []
+
+    if isinstance(result_query, str):
+        return result_query, pd.DataFrame(), []
+
+    if not result_query:
+        return f"No results for '{question}'. Try rephrasing. (⏱ {elapsed}s)", pd.DataFrame(), []
+
+    df = pd.DataFrame(result_query)
+    if sort_by and sort_by in df.columns:
+        df = df.sort_values(by=sort_by)
+
+
+    if "image_url" in df.columns:
+        columns_of_interest = ["image_url", "year","fashion_collectionLabel", "reference_URL"]
+        df = df[columns_of_interest]
+        # Create a gallery: each item is (image_url, metadata string)
+        gallery_items = []
+        for _, row in df.iterrows():
+            image_url = row.get("image_url")
+            if not image_url:
+                continue
+            # Caption from other fields
+            caption = " | ".join(f"{k}: {v}" for k, v in row.items() if k != "image_url" and pd.notnull(v))
+            gallery_items.append((image_url, caption))
+        return f"Query returned {len(gallery_items)} image(s) in {elapsed} seconds.", pd.DataFrame(), gallery_items
+
+    return f"Query returned a table with {len(df)} row(s) in {elapsed} seconds.", df, []
+
+
+
+
+def handle_image_query(text):
+    if not text:
+        return []
+
+    try:
+        records = main_text_retrieve_images(text)
+        print(f"Retrieved {len(records)} records for query: {text}")
+        print(records)
+    except Exception as e:
+        return [("https://via.placeholder.com/300x200?text=Error", f"Error: {e}")]
+
+    gallery_items = []
+    for item in records:
+        image_url = item.get("image_url")
+        if not image_url:
+            continue
+        # Build a simple caption from the remaining fields
+        caption = " | ".join(f"{k}: {v}" for k, v in item.items() if k != "image_url")
+        gallery_items.append((image_url, caption))
+
+    return gallery_items
+
+# --- UI --- #
+with gr.Blocks() as demo:
+    gr.Markdown("# 🧵 FashionDB Interface")
+
+
+    with gr.Tab("Structured Query"):
+        gr.Markdown("Ask FashionDB anything and view results with images + metadata.")
+
+        with gr.Row():
+            query_input = gr.Textbox(label="Your question")
+            sort_input = gr.Textbox(label="Sort by (optional column name)", placeholder="e.g. year")
+
+        query_submit = gr.Button("Submit")
+
+        query_text_output = gr.Textbox(label="Message", interactive=False)
+        query_table_output = gr.Dataframe(label="Tabular Result", interactive=False)
+        query_gallery_output = gr.Gallery(label="Image Gallery")
+        query_submit.click(
+            fn=handle_structured_query,
+            inputs=[query_input, sort_input],
+            outputs=[
+                query_text_output,
+                query_table_output,
+                query_gallery_output
+            ]
+        )
+
+    with gr.Tab("Image Retrieval"):
+        gr.Markdown("Search for similar fashion show images based on a text description.")
+        image_text = gr.Textbox(label="Describe the kind of images you're looking for")
+        image_submit = gr.Button("Find Images")
+        image_gallery = gr.Gallery(label="Retrieved Images")
+
+        image_submit.click(handle_image_query, inputs=image_text, outputs=image_gallery)
+
+demo.launch()

gradio_app1.py

@@ -0,0 +1,110 @@
+import gradio as gr
+
+# --- Handlers --- #
+from src1.generate_queries_alternative import main_generate_queries
+import time
+import pandas as pd
+
+
+
+
+def handle_structured_query(question, sort_by=""):
+    if not question:
+        return "Please ask something 🙂", pd.DataFrame(), []
+
+    try:
+        start = time.time()
+        result_query, sparql_query = main_generate_queries(question)
+        elapsed = round(time.time() - start, 2)
+    except Exception as e:
+        return f"⚠️ Query failed: {e}", pd.DataFrame(), []
+
+    if isinstance(result_query, str):
+        return result_query, pd.DataFrame(), []
+
+    if not result_query:
+        return f"No results for '{question}'. Try rephrasing. (⏱ {elapsed}s)", pd.DataFrame(), []
+
+    df = pd.DataFrame(result_query)
+    if sort_by and sort_by in df.columns:
+        df = df.sort_values(by=sort_by)
+
+
+    if "image_url" in df.columns:
+        columns_of_interest = ["image_url", "year","fashion_collectionLabel", "reference_URL"]
+        df = df[columns_of_interest]
+        # Create a gallery: each item is (image_url, metadata string)
+        gallery_items = []
+        for _, row in df.iterrows():
+            image_url = row.get("image_url")
+            if not image_url:
+                continue
+            # Caption from other fields
+            caption = " | ".join(f"{k}: {v}" for k, v in row.items() if k != "image_url" and pd.notnull(v))
+            gallery_items.append((image_url, caption))
+        return f"Query returned {len(gallery_items)} image(s) in {elapsed} seconds.", pd.DataFrame(), gallery_items
+
+    return f"Query returned a table with {len(df)} row(s) in {elapsed} seconds.", df, []
+
+
+
+
+
+from src1.visual_qa import main_text_retrieve_images
+
+def handle_image_query(text):
+    if not text:
+        return []
+
+    try:
+        records = main_text_retrieve_images(text)
+    except Exception as e:
+        return [("https://via.placeholder.com/300x200?text=Error", f"Error: {e}")]
+
+    gallery_items = []
+    for item in records:
+        image_url = item.get("image_url")
+        if not image_url:
+            continue
+        # Build a simple caption from the remaining fields
+        caption = " | ".join(f"{k}: {v}" for k, v in item.items() if k != "image_url")
+        gallery_items.append((image_url, caption))
+
+    return gallery_items
+
+# --- UI --- #
+with gr.Blocks() as demo:
+    gr.Markdown("# 🧵 FashionDB Interface")
+
+
+    with gr.Tab("Structured Query"):
+        gr.Markdown("Ask FashionDB anything and view results with images + metadata.")
+
+        with gr.Row():
+            query_input = gr.Textbox(label="Your question")
+            sort_input = gr.Textbox(label="Sort by (optional column name)", placeholder="e.g. start_year")
+
+        query_submit = gr.Button("Submit")
+
+        query_text_output = gr.Textbox(label="Message", interactive=False)
+        query_table_output = gr.Dataframe(label="Tabular Result", interactive=False)
+        query_gallery_output = gr.Gallery(label="Image Gallery")
+        query_submit.click(
+            fn=handle_structured_query,
+            inputs=[query_input, sort_input],
+            outputs=[
+                query_text_output,
+                query_table_output,
+                query_gallery_output
+            ]
+        )
+
+    with gr.Tab("Image Retrieval"):
+        gr.Markdown("Search for similar fashion show images based on a text description.")
+        image_text = gr.Textbox(label="Describe the kind of images you're looking for")
+        image_submit = gr.Button("Find Images")
+        image_gallery = gr.Gallery(label="Retrieved Images")
+
+        image_submit.click(handle_image_query, inputs=image_text, outputs=image_gallery)
+
+demo.launch( share=True)

search.py

@@ -0,0 +1,22 @@
+from sentence_transformers import SentenceTransformer
+from sklearn.metrics.pairwise import cosine_similarity
+import numpy as np
+
+# Use a compatible CLIP model
+model = SentenceTransformer("clip-ViT-B-32")
+
+def search_images_by_text(text, df, embeddings, top_k=30):
+    text_emb = model.encode([text])
+    filtered_embeddings = embeddings[df.index]
+    sims = cosine_similarity(text_emb, filtered_embeddings)[0]
+    top_indices = np.argsort(sims)[::-1][:top_k]
+    return df.iloc[top_indices]
+
+def get_similar_images(df, image_id, embeddings, top_k=5):
+    index = int(image_id)  # adjust based on your ID setup
+    query_emb = embeddings[index]
+    sims = cosine_similarity([query_emb], embeddings)[0]
+    top_indices = np.argsort(sims)[::-1][1:top_k+1]
+    return df.iloc[top_indices]
+
+