Upload 2 files

app.py

@@ -0,0 +1,184 @@
+import gradio as gr
+import json
+from gliner2 import GLiNER2
+from huggingface_hub import login
+import os
+
+# Get API key from environment variable
+hf_token = os.getenv("HF_TOKEN")
+
+# Authenticate with Hugging Face
+login(hf_token)
+
+# ——— Load model once ———
+model = GLiNER2.from_pretrained("fastino/gliner2-base-0207")
+
+def run_ner(text, types_csv, descs):
+    types = [t.strip() for t in types_csv.split(",") if t.strip()]
+    desc_map = {k: v for line in descs.split("\n") if ":" in line for k,v in [line.split(":",1)]}
+    inp = desc_map if desc_map else types
+    res = model.extract_entities(text=text, entity_types=inp, include_confidence=True)
+    return model.pretty_print_results(res, include_confidence=True)
+
+def run_class(text, task, labels_csv, descs, multi):
+    labels = [l.strip() for l in labels_csv.split(",") if l.strip()]
+    desc_map = {k: v for line in descs.split("\n") if ":" in line for k,v in [line.split(":",1)]}
+    inp = desc_map if desc_map else labels
+    tasks = {
+        task: {
+            "labels": list(inp.keys()) if isinstance(inp,dict) else inp,
+            "multi_label": multi,
+            **({"label_descriptions": inp} if isinstance(inp,dict) else {})
+        }
+    }
+    res = model.classify_text(text=text, tasks=tasks, include_confidence=True)
+    return model.pretty_print_results(res, include_confidence=True)
+
+def run_struct(text, struct_json):
+    try:
+        cfg = json.loads(struct_json)
+    except json.JSONDecodeError as e:
+        return f"❌ Invalid JSON: {e}"
+    res = model.extract_json(text=text, structures=cfg, include_confidence=True)
+    return model.pretty_print_results(res, include_confidence=True)
+
+# ——— Clean White Theme & Layout ———
+custom_css = """
+body {
+  background: #ffffff !important;
+  font-family: 'Helvetica Neue', sans-serif;
+  color: #333333;
+}
+.gradio-container {
+  max-width: 600px;
+  padding: 0;
+  background: #ffffff;
+}
+header, .logo, .subtitle {
+  border: none !important;
+  box-shadow: none !important;
+}
+.gradio-container * {
+  box-shadow: none !important;
+}
+.card {
+  background: #ffffff;
+  padding: 15px;
+}
+label {
+  color: #444444;
+  font-weight: 600;
+}
+.gr-textbox textarea,
+.gr-code,
+.gr-dropdown,
+.gr-checkbox,
+.gr-button {
+  background: #ffffff !important;
+  box-shadow: none !important;
+}
+.accordion-button {
+  border: none !important;
+  box-shadow: none !important;
+  font-weight: 500;
+}
+.gr-button.primary {
+  background: #5b8def;
+  color: #ffffff;
+}
+"""
+
+with gr.Blocks(theme=gr.themes.Base(), css=custom_css) as demo:
+    # Header
+    gr.HTML("""
+    <header style="text-align:center; padding:10px 0;">
+      <div class="logo" style="font-size:1.8rem; font-weight:700; color:#333333;">🎯 GLiNER2</div>
+      <div class="subtitle" style="font-size:0.85rem; color:#777777;">Compact • White Theme • Screenshot-Ready</div>
+    </header>
+    """)
+
+    with gr.Tabs():
+        # Structure Extraction Tab
+        with gr.TabItem("Hierarchical Structure Extraction"):
+            with gr.Row(elem_classes="card"):
+                with gr.Column(scale=2):
+                    txt3 = gr.Textbox(
+                        label="Input text", lines=3,
+                        value=(
+                            "The Acme Pro Laptop 15” features an Intel Core i7 processor, 16GB RAM, 512GB SSD, "
+                            "and a 15.6-inch 4K display. Priced at $1,499, it offers Wi-Fi 6, Bluetooth 5.2, and "
+                            "a backlit keyboard."
+                        )
+                    )
+                    struct3 = gr.Code(
+                        language="json", lines=7,
+                        label = "Schema",
+                        value=json.dumps({
+                            "product": [
+                                "name::str::Product name and model",
+                                "price::str::Product cost",
+                                "features::list::Key product features",
+                                "category::[electronics|software|hardware]::str"
+                            ]
+                        }, indent=2)
+                    )
+                    btn3 = gr.Button("Predict", variant="primary")
+                with gr.Column(scale=1):
+                    out3 = gr.Code(language="json", lines=8, label="Output")
+                btn3.click(run_struct, [txt3, struct3], out3)
+
+        # NER Tab
+        with gr.TabItem("Named Entity Recognition"):
+            with gr.Row(elem_classes="card"):
+                with gr.Column(scale=2):
+                    txt1 = gr.Textbox(
+                        label="Text", lines=4,
+                        value=(
+                            "Dr. Alice Smith, Chief Data Scientist at OpenAI, spoke at the AI Summit "
+                            "in San Francisco on June 12, 2025, about advancements in large-scale language "
+                            "models, ethical AI guidelines, and real-world GPT-4 Turbo applications."
+                        )
+                    )
+                    types1 = gr.Textbox(label="Types (csv)", value="person, title, organization, event, location, date, topic")
+                    with gr.Accordion("Descriptions (opt)", open=False):
+                        desc1 = gr.Textbox(lines=4, placeholder=(
+                            "person: Full names\n"
+                            "title: Roles\n"
+                            "organization: Companies\n"
+                            "event: Conferences\n"
+                            "location: Cities\n"
+                            "date: Temporal expressions"
+                        ))
+                    btn1 = gr.Button("Predict", variant="primary")
+                with gr.Column(scale=1):
+                    out1 = gr.Code(language="json", lines=8)
+                btn1.click(run_ner, [txt1, types1, desc1], out1)
+
+        # Classification Tab
+        with gr.TabItem("Text Classification"):
+            with gr.Row(elem_classes="card"):
+                with gr.Column(scale=2):
+                    txt2 = gr.Textbox(
+                        label="Text", lines=4,
+                        value=(
+                            "The Q2 2025 financial report shows a 15% revenue increase driven by cloud "
+                            "services, offset by a 12% rise in R&D costs. Overall sentiment is cautiously "
+                            "optimistic among stakeholders."
+                        )
+                    )
+                    task2 = gr.Textbox(label="Task", value="financial_sentiment")
+                    labs2 = gr.Textbox(label="Labels (csv)", value="positive, negative, neutral, mixed, uncertain")
+                    with gr.Accordion("Label Descriptions (opt)", open=False):
+                        desc2 = gr.Textbox(lines=3, placeholder=(
+                            "positive: Favorable outcomes\n"
+                            "negative: Concerns raised\n"
+                            "neutral: Balanced reporting"
+                        ))
+                    multi2 = gr.Checkbox(label="Multi-label?", value=True)
+                    btn2 = gr.Button("Predict", variant="primary")
+                with gr.Column(scale=1):
+                    out2 = gr.Code(language="json", lines=8)
+                btn2.click(run_class, [txt2, task2, labs2, desc2, multi2], out2)
+
+
+    demo.launch(share=False, width=600, height=300)

requirements.txt

@@ -0,0 +1 @@
+gliner2