Spaces:
Sleeping
Sleeping
emar
commited on
Commit
•
09070fb
1
Parent(s):
e0e7586
test
Browse files
app.py
CHANGED
@@ -13,46 +13,46 @@ PERSIST_DIR = './storage'
|
|
13 |
# Configure the settings
|
14 |
DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
15 |
|
16 |
-
@spaces.GPU
|
17 |
-
def setup():
|
18 |
-
|
19 |
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
|
29 |
|
30 |
-
# load the existing index
|
31 |
-
@spaces.GPU
|
32 |
-
def load_context():
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
|
38 |
|
39 |
|
40 |
-
@spaces.GPU
|
41 |
-
def chatbot_response(message, history):
|
42 |
-
|
43 |
-
|
44 |
|
45 |
-
setup()
|
46 |
-
query_engine = load_context()
|
47 |
|
48 |
|
49 |
-
iface = gr.ChatInterface(
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
)
|
56 |
|
57 |
-
if __name__ == "__main__":
|
58 |
-
|
|
|
13 |
# Configure the settings
|
14 |
DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
15 |
|
16 |
+
# @spaces.GPU
|
17 |
+
# def setup():
|
18 |
+
# Settings.embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-base-en-v1.5", device="cpu")
|
19 |
|
20 |
+
# Settings.llm = HuggingFaceLLM(
|
21 |
+
# model_name="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
|
22 |
+
# tokenizer_name="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
|
23 |
+
# context_window=2048,
|
24 |
+
# max_new_tokens=256,
|
25 |
+
# generate_kwargs={"temperature": 0.7, "top_k": 50, "top_p": 0.95},
|
26 |
+
# device_map="auto",
|
27 |
+
# )
|
28 |
|
29 |
|
30 |
+
# # load the existing index
|
31 |
+
# @spaces.GPU
|
32 |
+
# def load_context():
|
33 |
+
# storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
|
34 |
+
# index = load_index_from_storage(storage_context)
|
35 |
+
# query_engine = index.as_query_engine()
|
36 |
+
# return query_engine
|
37 |
|
38 |
|
39 |
|
40 |
+
# @spaces.GPU
|
41 |
+
# def chatbot_response(message, history):
|
42 |
+
# response = query_engine.query(message)
|
43 |
+
# return str(response)
|
44 |
|
45 |
+
# setup()
|
46 |
+
# query_engine = load_context()
|
47 |
|
48 |
|
49 |
+
# iface = gr.ChatInterface(
|
50 |
+
# fn=chatbot_response,
|
51 |
+
# title="UESP Lore Chatbot: CPU bound version of Phi-3-mini",
|
52 |
+
# description="Low quality and extremely slow version of the ones you can find on the github page.: https://github.com/emarron/UESP-lore I am not paying to have Llama3 on here.",
|
53 |
+
# examples=["Who is Zaraphus?"],
|
54 |
+
# cache_examples=True,
|
55 |
+
# )
|
56 |
|
57 |
+
# if __name__ == "__main__":
|
58 |
+
# iface.launch()
|