multi task app demo

app.py

@@ -1,64 +1,251 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+import json
+import logging
+from huggingface_hub import InferenceClient, InferenceApiError
+
+# ── CONFIG & SETUP ──────────────────────────────────────────────────────────
+API_TOKEN = os.getenv("HF_API_TOKEN")
+if not API_TOKEN:
+    raise ValueError("HF_API_TOKEN environment variable is not set.")
+
+CLIENT = InferenceClient(provider="hf-inference", api_key=API_TOKEN)
+
+# Configure logging
+logging.basicConfig(
+    format="%(asctime)s %(levelname)s %(message)s",
+    level=logging.INFO
 )
+logger = logging.getLogger(__name__)
+
+# Common timeout for HTTP calls (in seconds)
+REQUEST_TIMEOUT = 30
+
+
+# ── GENERIC CALL WRAPPER ────────────────────────────────────────────────────
+def safe_call(fn, *args, **kwargs):
+    """Run an inference call, catch errors and log them."""
+    try:
+        return fn(*args, **kwargs)
+    except InferenceApiError as e:
+        logger.error(f"Inference API error: {e}")
+        return {"error": f"Inference API error: {e}"}
+    except json.JSONDecodeError as e:
+        logger.error(f"JSON decode error: {e}")
+        return {"error": "Invalid JSON input."}
+    except Exception:
+        logger.exception("Unexpected error")
+        return {"error": "An unexpected error occurred. Please try again."}
+
+
+# ── TASK FUNCTIONS ──────────────────────────────────────────────────────────
+
+def asr_task(audio_path):
+    if not audio_path:
+        return "Please upload an audio file."
+    return safe_call(
+        CLIENT.automatic_speech_recognition,
+        audio_path,
+        model="openai/whisper-large-v3",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+def chat_task(messages_str):
+    if not messages_str.strip():
+        return "Enter messages in JSON format."
+    try:
+        messages = json.loads(messages_str)
+        if not isinstance(messages, list):
+            raise ValueError("Messages must be a JSON list.")
+    except Exception as e:
+        logger.warning(f"Invalid chat input: {e}")
+        return "Invalid input. Please provide a JSON list of `{role,content}` objects."
+
+    response = safe_call(
+        CLIENT.chat.completions.create,
+        model="gpt2",
+        messages=messages,
+        timeout=REQUEST_TIMEOUT
+    )
+    return response.choices[0].message if hasattr(response, "choices") else response
+
+
+def fill_mask_task(text):
+    if "[MASK]" not in text:
+        return "Your input must contain the token `[MASK]`."
+    return safe_call(
+        CLIENT.fill_mask,
+        text,
+        model="google-bert/bert-base-uncased",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+def qa_task(question, context):
+    if not question or not context:
+        return "Both question and context are required."
+    return safe_call(
+        CLIENT.question_answering,
+        question=question,
+        context=context,
+        model="deepset/roberta-base-squad2",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+def summarization_task(text):
+    if len(text.split()) < 5:
+        return "Please provide at least 5 words to summarize."
+    return safe_call(
+        CLIENT.summarization,
+        text,
+        model="facebook/bart-large-cnn",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+def text_gen_task(prompt):
+    if not prompt.strip():
+        return "Prompt cannot be empty."
+    resp = safe_call(
+        CLIENT.chat.completions.create,
+        model="gpt2",
+        messages=[{"role": "user", "content": prompt}],
+        timeout=REQUEST_TIMEOUT
+    )
+    return resp.choices[0].message if hasattr(resp, "choices") else resp
+
+
+def image_classification_task(image_path):
+    if not image_path:
+        return "Please upload an image."
+    return safe_call(
+        CLIENT.image_classification,
+        image_path,
+        model="Falconsai/nsfw_image_detection",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+def feature_extraction_task(text):
+    if not text.strip():
+        return "Input text is empty."
+    return safe_call(
+        CLIENT.feature_extraction,
+        text,
+        model="intfloat/multilingual-e5-large-instruct",
+        timeout=REQUEST_TIMEOUT
+    )
+
+
+# ── GRADIO INTERFACE ────────────────────────────────────────────────────────
+with gr.Blocks() as demo:
+    gr.Markdown("## 🚀 HF Inference Multi-Task Demo (CPU Space)")
+
+    with gr.Tabs():
+
+        with gr.TabItem("ASR"):
+            asr_input  = gr.Audio(source="upload", type="filepath", label="Upload Audio")
+            asr_output = gr.Textbox(label="Transcript")
+            # Example audio files (replace with actual files in your repo)
+            gr.Examples(
+                examples=["sample1.flac", "sample2.wav"],
+                inputs=asr_input,
+                outputs=asr_output
+            )
+            asr_input.change(asr_task, asr_input, asr_output)
+
+        with gr.TabItem("Chat (LLM)"):
+            chat_in  = gr.Textbox(
+                label="Messages (JSON list)",
+                placeholder='[{"role":"user","content":"Hello"}]',
+                lines=3
+            )
+            chat_out = gr.Textbox(label="Bot Reply")
+            gr.Examples(
+                examples=[
+                    '[{"role":"user","content":"What is AI?"}]',
+                    '[{"role":"system","content":"You are a helpful assistant."},'
+                    '{"role":"user","content":"Tell me a joke."}]'
+                ],
+                inputs=chat_in,
+                outputs=chat_out,
+                fn=chat_task
+            )
+
+        with gr.TabItem("Fill Mask"):
+            mask_in  = gr.Textbox(
+                label="Text with [MASK]",
+                placeholder="The capital of France is [MASK]."
+            )
+            mask_out = gr.JSON(label="Fill Mask Results")
+            gr.Examples(
+                examples=[
+                    "The Eiffel Tower is located in [MASK].",
+                    "Machine learning models are [MASK] and powerful."
+                ],
+                inputs=mask_in,
+                outputs=mask_out,
+                fn=fill_mask_task
+            )
+
+        with gr.TabItem("Q&A"):
+            qa_q   = gr.Textbox(label="Question")
+            qa_ctx = gr.Textbox(label="Context", lines=4)
+            qa_out = gr.Textbox(label="Answer")
+            gr.Examples(
+                examples=[
+                    ["Who wrote 'Pride and Prejudice'?", "Jane Austen was an English novelist known for ..."],
+                    ["What is photosynthesis?", "Photosynthesis is the process by which green plants ..."]
+                ],
+                inputs=[qa_q, qa_ctx],
+                outputs=qa_out,
+                fn=qa_task
+            )
+
+        with gr.TabItem("Summarization"):
+            sum_in  = gr.Textbox(label="Text to Summarize", lines=4)
+            sum_out = gr.Textbox(label="Summary")
+            gr.Examples(
+                examples=[
+                    "The Industrial Revolution began in Britain in the late 18th century..."
+                ],
+                inputs=sum_in,
+                outputs=sum_out,
+                fn=summarization_task
+            )
+
+        with gr.TabItem("Text Generation"):
+            gen_in  = gr.Textbox(label="Prompt", lines=2)
+            gen_out = gr.Textbox(label="Generated Text")
+            gr.Examples(
+                examples=[
+                    "Once upon a time in a mystical land",
+                    "In the future, humans will live on Mars because"
+                ],
+                inputs=gen_in,
+                outputs=gen_out,
+                fn=text_gen_task
+            )
+
+        with gr.TabItem("Image Classification"):
+            img_in  = gr.Image(type="filepath", label="Upload Image")
+            img_out = gr.JSON(label="Classes")
+            gr.Examples(
+                examples=["cat.jpg", "dog.png"],
+                inputs=img_in,
+                outputs=img_out,
+                fn=image_classification_task
+            )
 
+        with gr.TabItem("Feature Extraction"):
+            fe_in  = gr.Textbox(label="Input Text", lines=2)
+            fe_out = gr.Dataframe(label="Embeddings")
+            gr.Examples(
+                examples=["Machine learning is fascinating."],
+                inputs=fe_in,
+                outputs=fe_out,
+                fn=feature_extraction_task
+            )
 
-if __name__ == "__main__":
-    demo.launch()
+    demo.launch(server_name="0.0.0.0", server_port=7860)

requirements.txt

@@ -1 +1,2 @@
-huggingface_hub==0.25.2
+huggingface_hub==0.25.2
+gradio