File size: 4,457 Bytes
82f8f5f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
# app.py
import gradio as gr
from openai import OpenAI

# ----------------------- CONSTANTS ----------------------- #
SYSTEM_PROMPT = """
Given the research context, design an ablation study for the specified module or process.\nBegin the design with a clear statement of the research objective, followed by a detailed description of the experiment setup.\nDo not include the discussion of results or conclusions in the response, as the focus is solely on the experimental design.\nThe response should be within 300 words. Present the response in **Markdown** format (use headings, bold text, and bullet or numbered lists where appropriate).
""".strip()

# ----------------------- HELPERS ------------------------- #

def prepare_user_prompt(
    research_background: str,
    method: str,
    experiment_setup: str,
    experiment_results: str,
    module_name: str,
) -> str:
    """Craft the ‘user’ portion of the OpenAI chat based on form inputs."""
    research_background_block = f"### Research Background\n{research_background}\n"
    method_block = f"### Method Section\n{method}\n"
    experiment_block = (
        "### Main Experiment Setup\n"
        f"{experiment_setup}\n\n"
        "### Main Experiment Results\n"
        f"{experiment_results}\n"
    )

    return (
        "## Research Context\n"
        f"{research_background_block}{method_block}{experiment_block}\n\n"
        f"Design an **ablation study** about **{module_name}** based on the research context above."
    )


def generate_ablation_design(
    research_background,
    method,
    experiment_setup,
    experiment_results,
    module_name,
    api_key,
):
    """Combine inputs ➜ call OpenAI ➜ return the ablation-study design text (Markdown)."""
    # 1 ) validate the API key
    if not api_key or not api_key.startswith("sk-"):
        return "❌ **Please enter a valid OpenAI API key in the textbox above.**"

    # 2 ) build the chat conversation
    messages = [
        {"role": "system", "content": SYSTEM_PROMPT},
        {
            "role": "user",
            "content": prepare_user_prompt(
                research_background,
                method,
                experiment_setup,
                experiment_results,
                module_name,
            ),
        },
    ]

    # 3 ) call the model
    client = OpenAI(api_key=api_key)
    try:
        response = client.chat.completions.create(
            model="gpt-4o-mini",
            messages=messages,
            max_tokens=2048,
            temperature=1,
        )
        return response.choices[0].message.content.strip()
    except Exception as e:
        return f"⚠️  **OpenAI error:** {e}"

# ----------------------- UI LAYOUT ----------------------- #
with gr.Blocks(title="Ablation Study Designer") as demo:
    gr.Markdown(
        """
        # 🧪 Ablation Study Designer  \n
        Supply your study details below, then click **Generate** to receive a tailored ablation-study design rendered in Markdown.
        """
    )

    # API-key field (required)
    api_key = gr.Textbox(
        label="🔑 OpenAI API Key (required)",
        type="password",
        placeholder="sk-...",
    )

    research_background = gr.Textbox(
        label="Research Background", lines=6, placeholder="Describe the broader research context…"
    )
    method = gr.Textbox(
        label="Method Description", lines=6, placeholder="Summarize the method section…"
    )
    experiment_setup = gr.Textbox(
        label="Main Experiment – Setup", lines=6, placeholder="Datasets, hyper-parameters, etc."
    )
    experiment_results = gr.Textbox(
        label="Main Experiment – Results", lines=6, placeholder="Key quantitative or qualitative findings…"
    )
    module_name = gr.Textbox(
        label="Module / Process for Ablation", placeholder="e.g., Attention mechanism"
    )

    generate_btn = gr.Button("Generate Ablation Study Design")

    # Markdown output area — dynamic, renders Markdown
    output_md = gr.Markdown(value="", label="Ablation Study Design")

    generate_btn.click(
        fn=generate_ablation_design,
        inputs=[
            research_background,
            method,
            experiment_setup,
            experiment_results,
            module_name,
            api_key,
        ],
        outputs=output_md,
    )

# ----------------------- LAUNCH -------------------------- #
if __name__ == "__main__":
    demo.launch()