Update Generate prompt for leonardo #56

AliLee0923 · 2024-11-29T00:24:01Z

Created a function for generating detailed image prompts based on an array of words using the OpenAI gpt-3.5-turbo-instruct model. It creates a pictogram-style description for each word, determining whether the word represents an ACTION or OBJECT, and formats the prompt accordingly. (generatePromptForImageGeneration)
Created another function for combining the functionality of generating word suggestions and prompts for Leonardo AI. It fetches suggestions based on a given prompt, processes them into a list of detailed image prompts, and returns the results. (getPromptsForLenonardo)
Created a function for generating a detailed image prompt for a single word using the OpenAI API. This is a simplified and focused version of generatePromptForImageGeneration, handling only one word at a time. (generateAPromptForLeonardo)

…into prompt-leonardo

RodriSanchez1 · 2025-01-13T15:51:08Z

src/engine.ts

+  words: string[];
+}): Promise<Array<{ word: string; prompt: string }>> {
+  const completionRequestParams = {
+    model: "gpt-3.5-turbo-instruct",


Update the model to the 4o mini

RodriSanchez1 · 2025-01-13T17:06:35Z

src/engine.ts

+// A function to generate a prompt for generating images from Leonardo AI using gpt-4o-mini and provided template and words
+export async function generatePromptForImageGeneration({
+  words,
+}: {
+  words: string[];
+}): Promise<Array<{ word: string; prompt: string }>> {
+  const completionRequestParams = {
+    model: "gpt-4o-mini",
+    messages: [
+      {
+        role: "system",
+        content: `Create a detailed prompt to generate a pictogram for each word of the words array: '${words}'. 
+    First, determine if this is primarily an ACTION or OBJECT, then create a prompt following the appropriate template below.
+
+    For ACTIONS (verbs, activities):
+    - Show a figure actively performing the action
+    - Include clear motion indicators where appropriate
+    - Focus on the most recognizable moment of the action
+    - Use side view if it better shows the action
+    - Include minimal but necessary context elements
+
+    Style requirements:
+    - Bold black outlines
+    - Flat colors
+    - High contrast
+    - Centered composition
+    - White background
+    - Simple geometric shapes
+
+    Return only the prompt for each word, no explanations. Keep it under 100 words for each word. 
+    The returned template should be like this: 
+    word1: 'prompt',
+    word2: 'prompt', 
+    ...
+    wordN: 'prompt'`,
+      },
+      {
+        role: "user",
+        content: `Generate Prompts for ${words}`,
+      },
+    ],
+    temperature: 0,
+    max_tokens: 1500,
+  };
+
+  const response = await globalConfiguration.openAIInstance.createCompletion(
+    completionRequestParams
+  );
+  const promptText = response.data?.choices[0]?.text;
+  if (!promptText)
+    throw new Error("Error generating prompt for image generation");
+  try {
+    // Split the text by newlines and parse each line
+    const lines = promptText.split("\n").filter((line) => line.trim());
+    return lines.map((line) => {
+      const [word, ...promptParts] = line.split(":");
+      return {
+        word: word.trim(),
+        prompt: promptParts
+          .join(":")
+          .trim()
+          .replace(/^['"]|['"]$/g, ""), // Remove quotes if present
+      };
+    });
+  } catch (error) {
+    throw new Error("Error parsing image generation prompts: " + error);
+  }
+}
+
+export async function getPromptsForLenonardo({
+  prompt,
+  maxWords,
+  language,
+}: {
+  prompt: string;
+  maxWords: number;
+  language: string;
+}) {
+  const promptedWords = await getWordSuggestions({
+    prompt,
+    maxWords,
+    language,
+  });
+  const leonardoPrompts = await generatePromptForImageGeneration({
+    words: promptedWords,
+  });
+
+  return leonardoPrompts;
+}


I think we are not going to use this two functions. At least their are not being added to the return of the init function. Do you think we are going to use this? Also the generatePromptForImageGeneration it has a really different prompt content than the generateAPromptForLeonardo.
If you consider that we are going to use it, add it to the init return. Otherwise remove this functions from the PR 👍

AliLee0923 and others added 4 commits November 27, 2024 16:57

Update Generate prompt for leonardo

f1f3493

Update prompt for generateAPromptForLeonardo function

25a0e41

Merge branch 'main' of https://github.com/cboard-org/cboard-ai-engine …

b0a9781

…into prompt-leonardo

Update GPT model to 4o

a8d1755

RodriSanchez1 reviewed Jan 13, 2025

View reviewed changes

Update engine for new GPT model

38b2758

RodriSanchez1 reviewed Jan 13, 2025

View reviewed changes

Remove unnecesary functions

37bf686

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update Generate prompt for leonardo #56

Update Generate prompt for leonardo #56

AliLee0923 commented Nov 29, 2024

RodriSanchez1 Jan 13, 2025

RodriSanchez1 Jan 13, 2025

Update Generate prompt for leonardo #56

Are you sure you want to change the base?

Update Generate prompt for leonardo #56

Conversation

AliLee0923 commented Nov 29, 2024

RodriSanchez1 Jan 13, 2025

Choose a reason for hiding this comment

RodriSanchez1 Jan 13, 2025

Choose a reason for hiding this comment