Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/cake.jpg filter=lfs diff=lfs merge=lfs -text
+examples/gourmet_burger.jpg filter=lfs diff=lfs merge=lfs -text
+examples/pancakes_berries.jpg filter=lfs diff=lfs merge=lfs -text
+examples/pizza_board.jpg filter=lfs diff=lfs merge=lfs -text
+examples/sushi_nigiri.jpg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,13 +1,20 @@
 ---
-title: Olive Gemma
-emoji: 💻
-colorFrom: green
-colorTo: pink
+title: OliveGemma
+emoji: 🫒
+colorFrom: blue
+colorTo: red
 sdk: gradio
 sdk_version: 6.22.0
-python_version: '3.12'
 app_file: app.py
-pinned: false
+short_description: Mediterranean & European diet recognition VLM
+python_version: "3.12"
+startup_duration_timeout: 30m
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# OliveGemma 🫒
+
+OliveGemma is a 3B visual-language model for fine-grained food recognition, built on the PaliGemma-2-3B backbone with LoRA fine-tuning on 17,340 images across 216 Mediterranean & European dish categories.
+
+Upload a food image and ask a question — the model will identify the dish, list likely ingredients, describe visual evidence, and more.
+
+**Model:** [`JamesZar/OliveGemma-3B`](https://huggingface.co/JamesZar/OliveGemma-3B) · **Paper:** [OliveGemma: A 3 Billion VLM for Recognising the Mediterranean & European Diet](https://huggingface.co/papers/2608.03428) · **Code:** [GitHub](https://github.com/tsiokris/OliveGemma)

app.py

@@ -0,0 +1,114 @@
+import spaces  # MUST come before any CUDA-touching import
+import torch
+import gradio as gr
+from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
+
+MODEL_ID = "JamesZar/OliveGemma-3B"
+
+processor = AutoProcessor.from_pretrained(MODEL_ID)
+model = PaliGemmaForConditionalGeneration.from_pretrained(
+    MODEL_ID, torch_dtype=torch.bfloat16
+).to("cuda").eval()
+
+QUESTIONS = [
+    "What is the name of this dish?",
+    "What are the likely ingredients of this dish?",
+    "What visible ingredients can you see?",
+    "What visual evidence supports this dish?",
+    "How is this dish different from a visually similar one?",
+]
+
+
+@spaces.GPU(duration=60)
+def recognize(image, question: str, max_new_tokens: int = 64) -> str:
+    """Recognise a Mediterranean or European dish from an image and answer a question about it.
+
+    Args:
+        image: A food photograph.
+        question: What to ask the model about the food (dish name, ingredients, etc.).
+        max_new_tokens: Maximum number of new tokens to generate.
+    """
+    from PIL import Image
+
+    if image is None:
+        return "Please upload an image."
+    if not isinstance(image, Image.Image):
+        image = Image.open(image)
+    image = image.convert("RGB")
+
+    # PaliGemma prompt format used during training
+    prompt = f"<image>answer en {question}\n"
+    inputs = processor(text=prompt, images=image, return_tensors="pt").to(model.device)
+    in_len = inputs["input_ids"].shape[-1]
+
+    with torch.inference_mode():
+        out = model.generate(
+            **inputs,
+            max_new_tokens=int(max_new_tokens),
+            do_sample=False,
+        )
+
+    answer = processor.decode(out[0][in_len:], skip_special_tokens=True).strip()
+    return answer
+
+
+CSS = """
+#col-container { max-width: 1100px; margin: 0 auto; }
+.dark .gradio-container { color: var(--body-text-color); }
+"""
+
+with gr.Blocks(theme=gr.themes.Citrus(), css=CSS) as demo:
+    gr.Markdown(
+        "# OliveGemma 🫒\n"
+        "A 3B visual-language model for fine-grained Mediterranean & European food recognition. "
+        "Upload a food photo and ask about the dish name, ingredients, or visual evidence.\n\n"
+        "Model: [`JamesZar/OliveGemma-3B`](https://huggingface.co/JamesZar/OliveGemma-3B) · "
+        "Paper: [2608.03428](https://huggingface.co/papers/2608.03428) · "
+        "Code: [GitHub](https://github.com/tsiokris/OliveGemma)"
+    )
+
+    with gr.Column(elem_id="col-container"):
+        with gr.Row():
+            image_input = gr.Image(type="pil", label="Food image", scale=1)
+            with gr.Column(scale=1):
+                question_input = gr.Dropdown(
+                    choices=QUESTIONS,
+                    value=QUESTIONS[0],
+                    label="Question",
+                    interactive=True,
+                )
+                recognize_btn = gr.Button("Recognise", variant="primary")
+
+        output_text = gr.Textbox(label="Answer", lines=4, interactive=False)
+
+        with gr.Accordion("Advanced settings", open=False):
+            max_tokens = gr.Slider(
+                minimum=16, maximum=256, value=64, step=16,
+                label="Max new tokens",
+            )
+
+        gr.Examples(
+            examples=[
+                ["examples/pizza_board.jpg", QUESTIONS[0]],
+                ["examples/sushi_nigiri.jpg", QUESTIONS[0]],
+                ["examples/pancakes_berries.jpg", QUESTIONS[0]],
+                ["examples/gourmet_burger.jpg", QUESTIONS[1]],
+                ["examples/macarons.jpg", QUESTIONS[0]],
+                ["examples/cake.jpg", QUESTIONS[2]],
+            ],
+            inputs=[image_input, question_input],
+            outputs=output_text,
+            fn=recognize,
+            cache_examples=True,
+            cache_mode="lazy",
+        )
+
+    recognize_btn.click(
+        fn=recognize,
+        inputs=[image_input, question_input, max_tokens],
+        outputs=output_text,
+        api_name="recognize",
+    )
+
+if __name__ == "__main__":
+    demo.launch(mcp_server=True)

requirements.txt

@@ -0,0 +1,4 @@
+transformers
+accelerate
+pillow
+torchvision