|
|
|
|
|
|
|
|
def generate_intent_samples(dictionaries, count=50, source_language="en"): |
|
|
samples = [] |
|
|
|
|
|
intents = [ |
|
|
("assist", "1", "support"), |
|
|
("explain", "2", "clarity"), |
|
|
("analyze", "3", "depth"), |
|
|
("guide", "1", "direction"), |
|
|
("summarize", "1", "brevity") |
|
|
] |
|
|
|
|
|
for i in range(count): |
|
|
goal, urgency, focus = intents[i % len(intents)] |
|
|
|
|
|
user_text = "Help me understand this." |
|
|
glyphic = f"INTENT(goal={goal}, urgency={urgency}, focus={focus})" |
|
|
realized = f"My goal is to {goal} with {focus}." |
|
|
|
|
|
samples.append({ |
|
|
"input": user_text, |
|
|
"glyphic": glyphic, |
|
|
"output": realized, |
|
|
"intent": { |
|
|
"goal": goal, |
|
|
"urgency": urgency, |
|
|
"focus": focus |
|
|
}, |
|
|
"emotion": "neutral", |
|
|
"sensory": "none", |
|
|
"social": "alone" |
|
|
}) |
|
|
|
|
|
return samples |
|
|
|
|
|
|
