t-SNE
​
image Crawling tool
Step 1. ChatGPT API + Grasshopper
Set up the ChatGPT API in the Grasshopper in Rhino, which connects the OpenAI Server to Grasshopper so that users can use ChatGPT directly in the Grasshopper environment instead of the Web server.
​
url = "https://api.openai.com/v1/completions"
data = json.dumps({ "model": "text-davinci-003", "prompt": request, "temperature": 0.5, "max_tokens": 1024})
headers = { "Content-Type": "application/json", "Authorization": "Bearer {0}".format(api_key) }
Step 2. Text Prompt Input Setup
‘Loft’ is the design function to create a bench in this framework, so consider creating a feedback loop where the model output can be generated and further refinement should be included.
​
“Generate 10 sets of coordinates for a convex curve in the XZ plane, where the y-values are all 0.
Those generated points will be control points of the interpolated closed curve. Ensure that the points create a convex circle that avoids intersection lines. Only write coordinates not texts.”
Step 3. Generating Curves
Step 4. Loft
Step 5. Topology Optimization
Topology optimization tailors design for optimal material distribution within defined boundaries and loads. For a bench stand, one delineates the bench's dimensions and expected loads, like user weight. With set boundary conditions and material properties, the design is processed through specialized solvers. The outcome often resembles organic, bone-like structures, maximizing efficiency and aesthetics. However, post-optimization adjustments ensure manufacturability and practicality. Validated using tools like finite element analysis, the design is readied for real-world application. Topology optimization melds science and art, ensuring structural innovation in objects like benches.
Step 6. Segmentation for 3D Printing
