File size: 3,145 Bytes
235b1ba 1141e72 6a07e12 3b2955b 6a07e12 ef74b61 41bc4c0 5365751 fef5776 235b1ba 5365751 7a49b3f 5365751 d988bde 6099d18 522c4c0 d988bde f4f569e d988bde 0d31859 d988bde 44dee58 f715e9b 924fdc4 f715e9b baf17f8 0314f35 baf17f8 d3888c8 3fcd81d f715e9b 7b718cf 4ea7bf8 762a536 4ea7bf8 762a536 bcfe3e4 22c87f5 894f942 22c87f5 6ec3458 d988bde 235b1ba |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 |
import gradio as gr
from huggingface_hub import InferenceClient
from sentence_transformers import SentenceTransformer
import torch
import numpy as np
with open("knowledge.txt" , "r", encoding="utf-8") as f:
knowledge_base = f.read()
print("Knowledge base loaded.")
cleaned_text = knowledge_base.strip()
chunks = cleaned_text.split("\n")
cleaned_chunks = []
for chunk in chunks:
stripped_chunk = chunk.strip()
if stripped_chunk:
cleaned_chunks.append(stripped_chunk)
print(cleaned_chunks)
model = SentenceTransformer('all-MiniLM-L6-v2')
chunk_embeddings = model.encode(cleaned_chunks, convert_to_tensor=True)
print(chunk_embeddings)
def get_top_chunks(query):
query_embedding = model.encode(query, convert_to_tensor=True)
query_embedding_normalized = query_embedding / query_embedding.norm()
chunk_embeddings_normalized = chunk_embeddings / chunk_embeddings.norm(dim=1, keepdim=True)
similarities = torch.matmul(chunk_embeddings_normalized, query_embedding_normalized)
print(similarities)
top_indices = torch.topk(similarities, k=3).indices
print(top_indices)
top_chunks = []
for i in top_indices:
chunk = chunks[i]
top_chunks.append(chunk)
return top_chunks
client = InferenceClient("google/gemma-3-27b-it")
def respond(message,history):
messages = [{"role": "system" , "content": "Your name is BloomBot and you're a supportive and helpful chatbot catered towards teens ages 10-18. You give clear kid-appropiate explainations and keep your explainations to 10 sentences maximum."
}]
if history:
messages.extend(history)
messages.append({"role" : "user", "content" : message})
response = ""
for message in client.chat_completion(
messages,
max_tokens = 500,
stream=True,
#temperature = .2
top_p = .2
):
token = message.choices[0].delta.content
response += token
yield response
print(response)
theme = gr.themes.Ocean(
primary_hue="pink",
secondary_hue="pink",
neutral_hue="fuchsia"
)
def display_image():
return "BloomBotB (1).png"
with gr.Blocks (theme=theme) as chatbot:
gr.Image(display_image())
gr.ChatInterface(respond, type = "messages", #theme = gr.themes.Ocean(
#primary_hue="pink",
#secondary_hue="pink",
#neutral_hue="fuchsia"),
title = "Hi, I'm BloomBot!",
examples = ["What are the different types of period products? ",
"What are some vitamins that are good for teenage girls?",
"What should I know about puberty?"]
)
with gr.Tab("Resources"):
gr.Markdown("### Resources")
gr.HTML("""
<a href="https://drive.google.com/file/d/1_KNELAUDLLidwAT3fs2JBuO1yPgMGoDv/view" target="_blank">
<button style="font-size:16px;padding:10px 20px;margin-top:10px;">
📄 Period Tracker
</button>
</a>
""")
#with gr.Tab("Educational PDFs"):
# gr.Markdown("### 📘 Helpful Resources")
chatbot.launch(debug=True)
|