f

Dockerfile

@@ -0,0 +1,29 @@
+FROM nvidia/cuda:12.3.2-cudnn9-devel-ubuntu22.04
+
+# 1. System packages
+RUN apt-get update && apt-get install -y \
+    python3 python3-pip python3-dev build-essential ninja-build git \
+    && rm -rf /var/lib/apt/lists/*
+
+# 2. Upgrade pip
+RUN pip install --upgrade pip
+
+# 3. Install GPU-compatible Torch
+RUN pip install torch==2.1.0+cu121 -f https://download.pytorch.org/whl/cu121
+
+# 4. Install flash_attn
+RUN pip install flash_attn
+
+# 5. Install other Python libraries you need
+RUN pip install transformers gradio
+
+# 6. Copy your application code into the container
+WORKDIR /app
+COPY . /app
+
+# 7. Expose port 7860 (Gradio default)
+EXPOSE 7860
+
+# 8. Launch your app
+CMD ["python3", "app.py"]
+

README.md

@@ -1,10 +1,12 @@
 ---
-title: Autotrain
-emoji: 📈
-colorFrom: purple
+title: Autotrain Deepseek
+emoji: 💬
+colorFrom: yellow
 colorTo: purple
-sdk: docker
+sdk: gradio
+sdk_version: 5.0.1
+app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).

app.py

@@ -0,0 +1,54 @@
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+def load_model():
+    """
+    Load the DeepSeek-R1 model. 
+    Note: We rely on flash_attn, so this should now work
+    once PyTorch+CUDA and flash_attn are installed.
+    """
+    try:
+        model = AutoModelForCausalLM.from_pretrained(
+            "deepseek-ai/DeepSeek-R1", 
+            trust_remote_code=True
+        )
+        tokenizer = AutoTokenizer.from_pretrained(
+            "deepseek-ai/DeepSeek-R1", 
+            trust_remote_code=True
+        )
+        # Return a text-generation pipeline
+        return pipeline("text-generation", model=model, tokenizer=tokenizer)
+    except Exception as e:
+        return f"Model Loading Error: {e}"
+
+model_pipeline = load_model()
+
+def process_text(input_text):
+    """
+    Uses the loaded DeepSeek-R1 pipeline to generate text.
+    """
+    if isinstance(model_pipeline, str):
+        return f"Error: {model_pipeline}"  # If model_pipeline is an error string
+    try:
+        # Adjust parameters as desired
+        outputs = model_pipeline(input_text, max_length=200, num_return_sequences=1)
+        return outputs[0]["generated_text"]
+    except Exception as e:
+        return f"Inference Error: {e}"
+
+with gr.Blocks() as demo:
+    gr.Markdown(
+        "# DeepSeek-R1 Text Generator\n"
+        "Enter a prompt and generate text using the DeepSeek-R1 model."
+    )
+    input_box = gr.Textbox(
+        lines=5, label="Input Prompt", placeholder="Type your prompt here..."
+    )
+    generate_btn = gr.Button("Generate")
+    output_box = gr.Textbox(
+        lines=10, label="Generated Text", placeholder="Generated text appears here..."
+    )
+
+    generate_btn.click(fn=process_text, inputs=input_box, outputs=output_box)
+
+demo.launch()