Good day
We have used ChatGPT to generate ecosystems that show different dots and connections. This doesn’t work well - so we have implemented a second prompt to optimize the coordinates and visualization.
Now, we have prepared everything on the Web Premium version of GPT.
On the API we use the GPT 4 Turbo version.
Does the turbo version or generally the API not have access to different modules of the web version, like calculations, excel, programming, etc.? As the results are quite different using the API in this specific task.
Same when we calculate coordinates in between 3-4 locations.
This is our prompt - any idea how to get those packages working on the API?
In the input [EC7] you can find an input describing a node plot. Your job is to run the code and create another code with x,y coordinates of the nodes. Use abbreviations for the node names and document them in the beginning.
You must print all node location as coordinates in the output! There must be coordinates for all nodes in the output. There must be an x and y position for all nodes in the form ‘Full Node Name (X-Position, Y-Position)’ , otherwise you have to run the prompt again. Use the amada-Kawai layout to place the nodes for an optimal spacing (don’t just put them in a circle…).
Label the output ‘[CNLOC]’ (return this label before anything else and before all code!!!) and put a line break with whitespace after it. Don’t explain your task, don’t return an introduction or conclusion. " "[EC7]
python
import matplotlib.pyplot as plt
import networkx as nx
Mapping node names to identifiers
nodes = {
““ECS””: ““E-Commerce Shoppers””,
““ES””: ““Entrepreneurial Students””,
““HRSP””: ““HR Service Providers””,
““PRA””: ““PR Agencies””,
““RI””: ““Research Institutions””,
““LSP””: ““Legal Service Providers””,
““GE””: ““Government Entities””,
““SBO””: ““Small Business Owners””,
““MA””: ““Marketing Agencies””,
““EIE””: ““Educational Institutions for Entrepreneurs””,
““FSF””: ““Financial Services Firms””,
““E””: ““Entrepreneurs””,
““VCF””: ““Venture Capital Firms””,
““GSC””: ““Growth-Seeking Companies””,
““ECP””: ““E-Commerce Platforms””,
““TSP””: ““Technology Service Providers””,
““AIMIC””: ““AI-Based Market Insight Companies””,
““SI””: ““Startup Incubators””,
““AP””: ““Accelerator Programs””
}
Creating a network graph
G = nx.Graph()
Adding nodes
G.add_nodes_from(nodes.keys())
Adding edges
edges = [
(““ECP””, ““ECS””),
(““ES””, ““EIE””), (““ES””, ““SI””),
(““HRSP””, ““GSC””), (““HRSP””, ““GE””),
(““PRA””, ““MA””), (““PRA””, ““GSC””),
(““RI””, ““AIMIC””), (““RI””, ““EIE””), (““RI””, ““GE””),
(““LSP””, ““SI””), (““LSP””, ““AP””), (““LSP””, ““ECP””),
(““GE””, ““RI””), (““GE””, ““HRSP””), (““GE””, ““SBO””),
(““SBO””, ““SI””), (““SBO””, ““FSF””), (““SBO””, ““GE””), (““SBO””, ““VCF””),
(““MA””, ““AP””), (““MA””, ““ECP””), (““MA””, ““GSC””), (““MA””, ““PRA””),
(““EIE””, ““SI””), (““EIE””, ““ES””), (““EIE””, ““RI””), (““EIE””, ““E””), (““EIE””, ““GSC””),
(““FSF””, ““SBO””), (““FSF””, ““E””), (““FSF””, ““GSC””), (““FSF””, ““VCF””), (““FSF””, ““ECP””),
(““E””, ““EIE””), (““E””, ““SI””), (““E””, ““FSF””), (““E””, ““TSP””), (““E””, ““VCF””),
(““VCF””, ““AP””), (““VCF””, ““SBO””), (““VCF””, ““FSF””), (““VCF””, ““E””), (““VCF””, ““GSC””), (““VCF””, ““ECP””),
(““GSC””, ““HRSP””), (““GSC””, ““MA””), (““GSC””, ““TSP””), (““GSC””, ““FSF””), (““GSC””, ““VCF””), (““GSC””, ““E””),
(““ECP””, ““ECS””), (““ECP””, ““AIMIC””), (““ECP””, ““TSP””), (““ECP””, ““LSP””), (““ECP””, ““MA””), (““ECP””, ““VCF””),
(““TSP””, ““SI””), (““TSP””, ““AP””), (““TSP””, ““RI””), (““TSP””, ““ECP””), (““TSP””, ““GSC””), (““TSP””, ““E””), (““TSP””, ““AIMIC””),
(““AIMIC””, ““ECP””), (““AIMIC””, ““TSP””), (““AIMIC””, ““RI””), (““AIMIC””, ““E””), (““AIMIC””, ““GSC””), (““AIMIC””, ““VCF””),
(““SI””, ““EIE””), (““SI””, ““LSP””), (““SI””, ““TSP””), (““SI””, ““E””), (““SI””, ““SBO””), (““SI””, ““ES””), (““SI””, ““AP””), (““SI””, ““MA””),
(““AP””, ““SI””), (““AP””, ““TSP””), (““AP””, ““VCF””), (““AP””, ““LSP””), (““AP””, ““MA””), (““AP””, ““GSC””), (““AP””, ““E””), (““AP””, ““FSF””)
]
G.add_edges_from(edges)
Setting up the plot
plt.figure(figsize=(16, 9))
pos = nx.spring_layout(G, seed=42) # Using spring layout
Drawing the network
nx.draw(G, pos, with_labels=True, labels=nodes, node_color=““skyblue””, edge_color=““gray””, font_size=8, node_size=2500, alpha=0.6)
plt.title(““Network Map of Connections””)
plt.axis(‘off’) # Turn off the axis
plt.show()
This code snippet maps each entity in the network to a unique identifier, making the network
easier to manage and understand. The comments provide documentation for these mappings, ensuring clarity on the relationship between identifiers and their respective entities."