Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -82,8 +82,25 @@ retriever = PineconeHybridSearchRetriever(
|
|
82 |
alpha=0.5,
|
83 |
)
|
84 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
85 |
# Initialize LLM
|
86 |
-
llm = ChatPerplexity(temperature=0, pplx_api_key=GROQ_API_KEY, model="llama-3.1-sonar-large-128k-chat", max_tokens=512, max_retries=2)
|
87 |
|
88 |
# Initialize Reranker
|
89 |
# model = HuggingFaceCrossEncoder(model_name="BAAI/bge-reranker-base")
|
|
|
82 |
alpha=0.5,
|
83 |
)
|
84 |
|
85 |
+
|
86 |
+
from langchain_huggingface import HuggingFaceEndpoint,
|
87 |
+
ChatHuggingFace
|
88 |
+
|
89 |
+
llm = HuggingFaceEndpoint(
|
90 |
+
repo_id="meta-llama/Llama-3.3-70B-Instruct",
|
91 |
+
task="text-generation",
|
92 |
+
max_new_tokens=512,
|
93 |
+
do_sample=False,
|
94 |
+
repetition_penalty=1.03,
|
95 |
+
)
|
96 |
+
|
97 |
+
llm = ChatHuggingFace(llm=llm, verbose=True)
|
98 |
+
|
99 |
+
|
100 |
+
|
101 |
+
|
102 |
# Initialize LLM
|
103 |
+
# llm = ChatPerplexity(temperature=0, pplx_api_key=GROQ_API_KEY, model="llama-3.1-sonar-large-128k-chat", max_tokens=512, max_retries=2)
|
104 |
|
105 |
# Initialize Reranker
|
106 |
# model = HuggingFaceCrossEncoder(model_name="BAAI/bge-reranker-base")
|