|
|
|
|
|
import gradio as gr |
|
|
import pandas as pd |
|
|
import numpy as np |
|
|
from sentence_transformers import SentenceTransformer |
|
|
import faiss |
|
|
|
|
|
|
|
|
embed_model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2") |
|
|
|
|
|
|
|
|
data = pd.DataFrame({ |
|
|
"name": ["HowDidIDo", "Museotainment", "Movitr"], |
|
|
"tagline": ["Online evaluation platform", "PacMan & Louvre meet", "Crowdsourced video translation"], |
|
|
"description": [ |
|
|
"Public speaking, Presentation skills and interview practice", |
|
|
"Interactive AR museum tours", |
|
|
"Video translation with voice and subtitles" |
|
|
] |
|
|
}) |
|
|
|
|
|
|
|
|
data_vecs = embed_model.encode(data["description"].tolist()) |
|
|
faiss.normalize_L2(data_vecs) |
|
|
index = faiss.IndexFlatIP(data_vecs.shape[1]) |
|
|
index.add(data_vecs) |
|
|
|
|
|
def recommend(query, top_k=3): |
|
|
query_vec = embed_model.encode([query]) |
|
|
faiss.normalize_L2(query_vec) |
|
|
scores, idx = index.search(query_vec, top_k) |
|
|
results = data.iloc[idx[0]].copy() |
|
|
results["score"] = scores[0] |
|
|
return results[["name", "tagline", "description", "score"]] |
|
|
|
|
|
def generate_slogan(query_text, neighbors_df=None, n_samples=16): |
|
|
ctx = _neighbor_context(neighbors_df) |
|
|
prompt = ( |
|
|
"You are a creative brand copywriter. Write short, original, memorable startup slogans (max 8 words). |
|
|
" |
|
|
"Forbidden words: app, assistant, platform, solution, system, marketplace, AI, machine learning, augmented reality, virtual reality, decentralized, empower. |
|
|
" |
|
|
"Focus on clear benefits and vivid verbs. Do not copy the description. Return ONLY a list, one slogan per line. |
|
|
|
|
|
" |
|
|
"Good Examples: |
|
|
" |
|
|
"Description: AI assistant for doctors to prioritize patient cases |
|
|
" |
|
|
"Slogan: Less Guessing. More Healing. |
|
|
|
|
|
" |
|
|
"Description: Payments for small online stores |
|
|
" |
|
|
"Slogan: Built to Grow with Your Cart. |
|
|
|
|
|
" |
|
|
"Description: Neurotech headset to boost focus |
|
|
" |
|
|
"Slogan: Train Your Brain to Win. |
|
|
|
|
|
" |
|
|
) |
|
|
if ctx: |
|
|
prompt += f"Similar taglines (style only): |
|
|
{ctx} |
|
|
|
|
|
" |
|
|
prompt += f"Description: {query_text} |
|
|
Slogans:" |
|
|
|
|
|
input_ids = GEN_TOK(prompt, return_tensors="pt").input_ids.to(DEVICE) |
|
|
outputs = GEN_MODEL.generate( |
|
|
input_ids, |
|
|
max_new_tokens=24, |
|
|
do_sample=True, |
|
|
top_k=60, |
|
|
top_p=0.92, |
|
|
temperature=1.2, |
|
|
num_return_sequences=n_samples, |
|
|
repetition_penalty=1.08 |
|
|
) |
|
|
|
|
|
raw_cands = [GEN_TOK.decode(o, skip_special_tokens=True) for o in outputs] |
|
|
|
|
|
cand_set = set() |
|
|
for txt in raw_cands: |
|
|
for line in txt.split(" |
|
|
"): |
|
|
s = _clean_slogan(line) |
|
|
if not s: |
|
|
continue |
|
|
if len(s.split()) < 2 or len(s.split()) > 8: |
|
|
continue |
|
|
if _is_blocked_slogan(s): |
|
|
continue |
|
|
cand_set.add(_titlecase_soft(s)) |
|
|
|
|
|
if not cand_set: |
|
|
return _clean_slogan(GEN_TOK.decode(outputs[0], skip_special_tokens=True)) |
|
|
|
|
|
scored = _score_candidates(query_text, sorted(cand_set), neighbors_df) |
|
|
if not scored: |
|
|
return _clean_slogan(GEN_TOK.decode(outputs[0], skip_special_tokens=True)) |
|
|
|
|
|
scored.sort(key=lambda x: x[1], reverse=True) |
|
|
return scored[0][0] |
|
|
|
|
|
|
|
|
def pipeline(user_input): |
|
|
recs = recommend(user_input, top_k=3) |
|
|
slogan = generate_slogan(user_input) |
|
|
recs = recs.reset_index(drop=True) |
|
|
recs.loc[len(recs)] = ["Generated Slogan", slogan, user_input, np.nan] |
|
|
return recs |
|
|
|
|
|
examples = [ |
|
|
"AI coach for improving public speaking skills", |
|
|
"Augmented reality app for interactive museum tours", |
|
|
"Voice-controlled task manager for remote teams", |
|
|
"Machine learning system for predicting crop yields", |
|
|
"Platform for AI-assisted interior design suggestions" |
|
|
] |
|
|
|
|
|
demo = gr.Interface( |
|
|
fn=pipeline, |
|
|
inputs=gr.Textbox(label="Enter a startup description"), |
|
|
outputs=gr.Dataframe(headers=["Name", "Tagline", "Description", "Score"]), |
|
|
examples=examples, |
|
|
title="SloganAI – Startup Recommendation & Slogan Generator", |
|
|
description="Enter a startup idea and get top-3 similar startups + 1 generated slogan." |
|
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.launch() |
|
|
|
