fix(gradio): use rock-solid ChatInterface to eliminate UI validation errors

colab_demo_olmo2_astro_logic.ipynb

@@ -8,17 +8,18 @@
     "### Interactive Colab GPU Demo with Gradio (First-Gen Domain Foundation)\n",
     "\n",
     "**Architect & Creator:** Shreyansh Singh \u00b7 [ExperimentLab](https://github.com/shreyansh001boy-tech) \u00b7 [Hugging Face Repository](https://huggingface.co/shreyansh12183/olmo2-7b-astro-logic)  \n",
+    "**Architecture Lineage:** `Generation 1 \u00b7 Domain Specialist`  \n",
     "**Certified Benchmark:** `Certified Astrophysical Physics & Formal Proofs | General Relativity & Symbolic Logic`  \n",
-    "**Base Foundation:** `allenai/OLMo-2-1124-7B-Instruct` | **Adapter:** `shreyansh12183/olmo2-7b-astro-logic`  \n",
+    "**Model Identifier:** [`shreyansh12183/olmo2-7b-astro-logic`](https://huggingface.co/shreyansh12183/olmo2-7b-astro-logic)  \n",
     "**License:** CC BY-NC 4.0 (Non-Commercial Sovereign License)\n",
     "\n",
     "---\n",
     "\n",
     "### \ud83d\ude80 Running This Demo on Google Colab:\n",
-    "1. Ensure your runtime is set to **T4 GPU** (`Runtime -> Change runtime type -> T4 GPU`).\n",
-    "2. Run **Cell 1** to install dependencies (`transformers`, `bitsandbytes`, `peft`, `gradio`).\n",
-    "3. Run **Cell 2** to load the 7B foundation weights in 4-bit NF4 into Colab T4 GPU VRAM (~5.5 GB).\n",
-    "4. Run **Cell 3** to launch the interactive Gradio web application with a public `https://*.gradio.live` link!\n",
+    "1. Ensure runtime is set to **T4 GPU** (`Runtime -> Change runtime type -> T4 GPU`).\n",
+    "2. Run **Cell 1** to install required libraries (`transformers`, `bitsandbytes`, `peft`, `gradio`).\n",
+    "3. Run **Cell 2** to load model weights in 4-bit NF4 into Colab GPU VRAM.\n",
+    "4. Run **Cell 3** to launch the interactive Gradio web app and receive your public `https://*.gradio.live` link!\n",
     "\n"
    ]
   },
@@ -28,7 +29,7 @@
    "execution_count": null,
    "outputs": [],
    "source": [
-    "# Cell 1: Install Required Libraries\n!pip install -q torch transformers bitsandbytes peft accelerate gradio\nprint('\u2713 Dependencies installed successfully!')\n"
+    "# Cell 1: Install Required Libraries\n!pip install -q \"gradio>=4.44.1\" torch transformers bitsandbytes peft accelerate\nprint('\u2713 Dependencies installed successfully!')\n"
    ]
   },
   {
@@ -37,7 +38,7 @@
    "execution_count": null,
    "outputs": [],
    "source": [
-    "# Cell 2: Load 7B Base Foundation & LoRA Adapter in 4-bit NF4\nimport torch\nfrom transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, TextIteratorStreamer\nfrom peft import PeftModel\nimport threading\n\nBASE_ID = \"allenai/OLMo-2-1124-7B-Instruct\"\nADAPTER_ID = \"shreyansh12183/olmo2-7b-astro-logic\"\n\nprint(f\"Loading Base Foundation: {BASE_ID} in 4-bit NF4...\")\nbnb_config = BitsAndBytesConfig(\n    load_in_4bit=True,\n    bnb_4bit_quant_type=\"nf4\",\n    bnb_4bit_compute_dtype=torch.float16,\n    bnb_4bit_use_double_quant=True\n)\n\ntokenizer = AutoTokenizer.from_pretrained(BASE_ID, trust_remote_code=True)\nif tokenizer.pad_token is None:\n    tokenizer.pad_token = tokenizer.eos_token\n\nbase_model = AutoModelForCausalLM.from_pretrained(\n    BASE_ID,\n    quantization_config=bnb_config,\n    device_map=\"auto\",\n    torch_dtype=torch.float16,\n    trust_remote_code=True\n)\n\nprint(f\"Attaching Fine-Tuned LoRA Adapter: {ADAPTER_ID}...\")\nmodel = PeftModel.from_pretrained(base_model, ADAPTER_ID)\nmodel.eval()\nprint(f\"\u2713 {ADAPTER_ID} successfully loaded in Colab GPU VRAM (~5.5 GB)!\")\n"
+    "# Cell 2: Load Base Model & LoRA Adapter in 4-bit NF4\nimport torch\nfrom transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, TextIteratorStreamer\nfrom peft import PeftModel\nimport threading\n\nBASE_ID = \"allenai/OLMo-2-1124-7B-Instruct\"\nADAPTER_ID = \"shreyansh12183/olmo2-7b-astro-logic\"\n\nprint(f\"Loading Base Foundation: {BASE_ID} in 4-bit NF4...\")\nbnb_config = BitsAndBytesConfig(\n    load_in_4bit=True,\n    bnb_4bit_quant_type=\"nf4\",\n    bnb_4bit_compute_dtype=torch.float16,\n    bnb_4bit_use_double_quant=True\n)\n\ntokenizer = AutoTokenizer.from_pretrained(BASE_ID, trust_remote_code=True)\nif tokenizer.pad_token is None:\n    tokenizer.pad_token = tokenizer.eos_token\n\nbase_model = AutoModelForCausalLM.from_pretrained(\n    BASE_ID,\n    quantization_config=bnb_config,\n    device_map=\"auto\",\n    torch_dtype=torch.float16,\n    trust_remote_code=True\n)\n\nprint(f\"Attaching Fine-Tuned LoRA Adapter: {ADAPTER_ID}...\")\nmodel = PeftModel.from_pretrained(base_model, ADAPTER_ID)\nmodel.eval()\nprint(f\"\u2713 {ADAPTER_ID} successfully loaded in Colab GPU VRAM!\")\n"
    ]
   },
   {
@@ -46,7 +47,7 @@
    "execution_count": null,
    "outputs": [],
    "source": [
-    "# Cell 3: Launch Live Gradio Web Interface\nimport gradio as gr\n\nSYSTEM_PROMPT = \"You are OLMo-2 7B Astro Logic, an astrophysics and formal symbolic reasoning foundation model architected and fine-tuned by Shreyansh Singh (ExperimentLab). Solve complex celestial mechanics, general relativistic field equations, stellar nucleosynthesis problems, and rigorous formal symbolic logic derivations.\"\nEXAMPLES = [\"Derive the Schwarzschild metric geodesic equations and calculate the perihelion precession of Mercury.\", \"Explain the proton-proton chain vs CNO cycle in stellar nucleosynthesis and derive the Gamow peak energy window.\", \"Formulate the cosmological Friedmann equations from the FLRW metric and solve for a matter-dominated flat universe.\", \"Provide a formal symbolic natural deduction proof for the law of excluded middle in propositional logic.\"]\n\ndef generate_response(message, history):\n    messages = [{\"role\": \"system\", \"content\": SYSTEM_PROMPT}]\n    for user_msg, bot_msg in history:\n        messages.append({\"role\": \"user\", \"content\": user_msg})\n        if bot_msg:\n            messages.append({\"role\": \"assistant\", \"content\": bot_msg})\n    messages.append({\"role\": \"user\", \"content\": message})\n    \n    prompt_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)\n    inputs = tokenizer(prompt_text, return_tensors=\"pt\").to(model.device)\n    \n    streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)\n    gen_kwargs = dict(\n        input_ids=inputs.input_ids,\n        attention_mask=inputs.attention_mask,\n        streamer=streamer,\n        max_new_tokens=1024,\n        temperature=0.3,\n        top_p=0.9,\n        repetition_penalty=1.1,\n        do_sample=True\n    )\n    \n    thread = threading.Thread(target=model.generate, kwargs=gen_kwargs)\n    thread.start()\n    \n    partial_text = \"\"\n    for new_text in streamer:\n        partial_text += new_text\n        yield partial_text\n\ncustom_css = \"\"\"\n#header { text-align: center; margin-bottom: 15px; }\n#badge { background: #1e293b; color: #38bdf8; padding: 6px 14px; border-radius: 9999px; font-weight: 600; display: inline-block; font-size: 13px; }\n\"\"\"\n\nwith gr.Blocks(title=\"\ud83d\udd2d OLMo-2 7B Astro Logic Specialist\", css=custom_css, theme=gr.themes.Soft()) as demo:\n    with gr.Column(elem_id=\"header\"):\n        gr.Markdown(f\"# \ud83d\udd2d OLMo-2 7B Astro Logic Specialist\")\n        gr.Markdown(f\"### Interactive Colab GPU Demo with Gradio (First-Gen Domain Foundation)\")\n        gr.Markdown(f\"<span id='badge'>Certified: Certified Astrophysical Physics & Formal Proofs | General Relativity & Symbolic Logic</span>\")\n        gr.Markdown(f\"**Base Foundation:** `allenai/OLMo-2-1124-7B-Instruct` \u00b7 **Adapter:** [`shreyansh12183/olmo2-7b-astro-logic`](https://huggingface.co/shreyansh12183/olmo2-7b-astro-logic) \u00b7 **Architect:** Shreyansh Singh\")\n        \n    chatbot = gr.Chatbot(height=520, show_label=False, show_copy_button=True)\n    with gr.Row():\n        msg_input = gr.Textbox(placeholder=\"Enter your prompt here or click any of the curated questions below...\", scale=8, show_label=False)\n        send_btn = gr.Button(\"Submit\", variant=\"primary\", scale=2)\n        clear_btn = gr.Button(\"Clear Chat\", scale=1)\n        \n    gr.Markdown(\"### \ud83c\udfaf Curated Evaluation & Demo Questions (Click to Test):\")\n    gr.Examples(examples=EXAMPLES, inputs=msg_input)\n    \n    def user(user_message, history):\n        return \"\", history + [[user_message, None]]\n\n    def bot(history):\n        user_message = history[-1][0]\n        history_prior = history[:-1]\n        history[-1][1] = \"\"\n        for chunk in generate_response(user_message, history_prior):\n            history[-1][1] = chunk\n            yield history\n\n    msg_input.submit(user, [msg_input, chatbot], [msg_input, chatbot], queue=False).then(\n        bot, chatbot, chatbot\n    )\n    send_btn.click(user, [msg_input, chatbot], [msg_input, chatbot], queue=False).then(\n        bot, chatbot, chatbot\n    )\n    clear_btn.click(lambda: None, None, chatbot, queue=False)\n\nprint(\"\\n\" + \"=\"*75)\nprint(\"\ud83d\ude80 Launching Gradio Web App with Public Share Link...\")\nprint(\"=\"*75 + \"\\n\")\ndemo.queue()\ndemo.launch(share=True, debug=True)\n"
+    "# Cell 3: Launch Live Gradio Web Interface (Rock-Solid ChatInterface)\nimport gradio as gr\nimport threading\nfrom transformers import TextIteratorStreamer\n\nSYSTEM_PROMPT = \"You are OLMo-2 7B Astro Logic, an astrophysics and formal symbolic reasoning foundation model architected and fine-tuned by Shreyansh Singh (ExperimentLab). Solve complex celestial mechanics, general relativistic field equations, stellar nucleosynthesis problems, and rigorous formal symbolic logic derivations.\"\nEXAMPLES = [\"Derive the Schwarzschild metric geodesic equations and calculate the perihelion precession of Mercury.\", \"Explain the proton-proton chain vs CNO cycle in stellar nucleosynthesis and derive the Gamow peak energy window.\", \"Formulate the cosmological Friedmann equations from the FLRW metric and solve for a matter-dominated flat universe.\", \"Provide a formal symbolic natural deduction proof for the law of excluded middle in propositional logic.\"]\nTITLE = \"\\ud83d\\udd2d OLMo-2 7B Astro Logic Specialist\"\nDESCRIPTION = \"**Generation:** `Generation 1 \\u00b7 Domain Specialist` \\u00b7 **Model:** [`shreyansh12183/olmo2-7b-astro-logic`](https://huggingface.co/shreyansh12183/olmo2-7b-astro-logic) \\u00b7 **Certified:** `Certified Astrophysical Physics & Formal Proofs | General Relativity & Symbolic Logic`\"\n\ndef predict(message, history):\n    messages = [{\"role\": \"system\", \"content\": SYSTEM_PROMPT}]\n    for item in history:\n        if isinstance(item, dict):\n            role = item.get(\"role\", \"user\")\n            content = item.get(\"content\", \"\")\n            if content:\n                messages.append({\"role\": role, \"content\": content})\n        elif isinstance(item, (list, tuple)) and len(item) == 2:\n            u_msg, a_msg = item[0], item[1]\n            if u_msg:\n                messages.append({\"role\": \"user\", \"content\": str(u_msg)})\n            if a_msg:\n                messages.append({\"role\": \"assistant\", \"content\": str(a_msg)})\n    messages.append({\"role\": \"user\", \"content\": message})\n    \n    prompt_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)\n    inputs = tokenizer(prompt_text, return_tensors=\"pt\").to(model.device)\n    \n    streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)\n    gen_kwargs = dict(\n        input_ids=inputs.input_ids,\n        attention_mask=inputs.attention_mask,\n        streamer=streamer,\n        max_new_tokens=1024,\n        temperature=0.3,\n        top_p=0.9,\n        repetition_penalty=1.1,\n        do_sample=True\n    )\n    thread = threading.Thread(target=model.generate, kwargs=gen_kwargs)\n    thread.start()\n    \n    partial_text = \"\"\n    for new_text in streamer:\n        partial_text += new_text\n        yield partial_text\n\ndemo = gr.ChatInterface(\n    fn=predict,\n    title=TITLE,\n    description=DESCRIPTION,\n    examples=EXAMPLES,\n)\n\nprint(\"\\n\" + \"=\"*75)\nprint(\"\ud83d\ude80 Launching Gradio Web App with Public Share Link...\")\nprint(\"=\"*75 + \"\\n\")\ndemo.launch(share=True)\n"
    ]
   }
  ],