Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,39 +1,39 @@
|
|
1 |
import os
|
2 |
-
|
|
|
3 |
from sentence_transformers import SentenceTransformer
|
4 |
from sklearn.metrics.pairwise import cosine_similarity
|
5 |
import numpy as np
|
6 |
import pandas as pd
|
7 |
-
import gradio as gr
|
8 |
|
9 |
# Load pre-trained Sentence Transformer model
|
10 |
-
|
11 |
|
12 |
# Load questions and answers from the CSV file
|
13 |
df = pd.read_csv('combined_questions_and_answers.csv')
|
14 |
|
15 |
# Encode all questions in the dataset
|
16 |
-
question_embeddings =
|
17 |
|
18 |
# Hugging Face API details for Meta-Llama 3B
|
19 |
-
|
20 |
-
if not
|
21 |
raise ValueError("Hugging Face API key not found in environment variables. Please set the HUGGINGFACE_API_KEY environment variable.")
|
22 |
|
23 |
-
|
|
|
|
|
24 |
|
25 |
# Function to refine and translate text using Meta-Llama 3B
|
26 |
def refine_text(prompt):
|
27 |
-
|
28 |
-
|
29 |
-
]
|
30 |
-
response = pipe(messages)
|
31 |
-
return response[0]['generated_text']
|
32 |
|
33 |
# Function to find the most similar question and provide the answer
|
34 |
def get_answer(user_question, threshold=0.30):
|
35 |
# Encode the user question
|
36 |
-
user_embedding =
|
37 |
|
38 |
# Calculate cosine similarities
|
39 |
similarities = cosine_similarity([user_embedding], question_embeddings)
|
|
|
1 |
import os
|
2 |
+
import gradio as gr
|
3 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
4 |
from sentence_transformers import SentenceTransformer
|
5 |
from sklearn.metrics.pairwise import cosine_similarity
|
6 |
import numpy as np
|
7 |
import pandas as pd
|
|
|
8 |
|
9 |
# Load pre-trained Sentence Transformer model
|
10 |
+
model_sentence_transformer = SentenceTransformer('LaBSE')
|
11 |
|
12 |
# Load questions and answers from the CSV file
|
13 |
df = pd.read_csv('combined_questions_and_answers.csv')
|
14 |
|
15 |
# Encode all questions in the dataset
|
16 |
+
question_embeddings = model_sentence_transformer.encode(df['Question'].tolist())
|
17 |
|
18 |
# Hugging Face API details for Meta-Llama 3B
|
19 |
+
HF_TOKEN = os.environ.get("HUGGINGFACE_API_KEY", None)
|
20 |
+
if not HF_TOKEN:
|
21 |
raise ValueError("Hugging Face API key not found in environment variables. Please set the HUGGINGFACE_API_KEY environment variable.")
|
22 |
|
23 |
+
# Load the tokenizer and model
|
24 |
+
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct")
|
25 |
+
model = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct", device_map="auto")
|
26 |
|
27 |
# Function to refine and translate text using Meta-Llama 3B
|
28 |
def refine_text(prompt):
|
29 |
+
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
|
30 |
+
outputs = model.generate(**inputs, max_new_tokens=50)
|
31 |
+
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
|
|
|
|
32 |
|
33 |
# Function to find the most similar question and provide the answer
|
34 |
def get_answer(user_question, threshold=0.30):
|
35 |
# Encode the user question
|
36 |
+
user_embedding = model_sentence_transformer.encode(user_question)
|
37 |
|
38 |
# Calculate cosine similarities
|
39 |
similarities = cosine_similarity([user_embedding], question_embeddings)
|