Spaces:
Paused
Paused
dev/check-gpu-for-rinna (#10)
Browse files- unavailable rinnna at no GPU (77ef434fe49c5896ce25aba5f08422384004d16d)
app.py
CHANGED
@@ -25,14 +25,17 @@ E5_EMBEDDINGS = HuggingFaceEmbeddings(
|
|
25 |
encode_kwargs=E5_ENCODE_KWARGS,
|
26 |
)
|
27 |
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
|
|
|
|
|
|
36 |
|
37 |
|
38 |
def _get_config_and_embeddings(collection_name: str | None) -> tuple:
|
@@ -48,14 +51,17 @@ def _get_config_and_embeddings(collection_name: str | None) -> tuple:
|
|
48 |
|
49 |
|
50 |
def _get_rinna_llm(temperature: float):
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
|
|
|
|
|
|
59 |
return llm
|
60 |
|
61 |
|
@@ -64,7 +70,7 @@ def _get_llm_model(
|
|
64 |
temperature: float,
|
65 |
):
|
66 |
if model_name is None:
|
67 |
-
model = "
|
68 |
elif model_name == "rinna":
|
69 |
model = "rinna"
|
70 |
elif model_name == "GPT-3.5":
|
@@ -157,12 +163,17 @@ def main(
|
|
157 |
return result["result"], html
|
158 |
|
159 |
|
|
|
|
|
|
|
|
|
|
|
160 |
nvdajp_book_qa = gr.Interface(
|
161 |
fn=main,
|
162 |
inputs=[
|
163 |
gr.Textbox(label="query"),
|
164 |
gr.Radio(["E5", "OpenAI"], label="Embedding", info="選択なしで「E5」を使用"),
|
165 |
-
gr.Radio(
|
166 |
gr.Radio(
|
167 |
["All", "ja-book", "ja-nvda-user-guide", "en-nvda-user-guide"],
|
168 |
label="絞り込み",
|
|
|
25 |
encode_kwargs=E5_ENCODE_KWARGS,
|
26 |
)
|
27 |
|
28 |
+
if torch.cuda.is_available():
|
29 |
+
RINNA_MODEL_NAME = "rinna/bilingual-gpt-neox-4b-instruction-ppo"
|
30 |
+
RINNA_TOKENIZER = AutoTokenizer.from_pretrained(RINNA_MODEL_NAME, use_fast=False)
|
31 |
+
RINNA_MODEL = AutoModelForCausalLM.from_pretrained(
|
32 |
+
RINNA_MODEL_NAME,
|
33 |
+
load_in_8bit=True,
|
34 |
+
torch_dtype=torch.float16,
|
35 |
+
device_map="auto",
|
36 |
+
)
|
37 |
+
else:
|
38 |
+
RINNA_MODEL = None
|
39 |
|
40 |
|
41 |
def _get_config_and_embeddings(collection_name: str | None) -> tuple:
|
|
|
51 |
|
52 |
|
53 |
def _get_rinna_llm(temperature: float):
|
54 |
+
if RINNA_MODEL is not None:
|
55 |
+
pipe = pipeline(
|
56 |
+
"text-generation",
|
57 |
+
model=RINNA_MODEL,
|
58 |
+
tokenizer=RINNA_TOKENIZER,
|
59 |
+
max_new_tokens=1024,
|
60 |
+
temperature=temperature,
|
61 |
+
)
|
62 |
+
llm = HuggingFacePipeline(pipeline=pipe)
|
63 |
+
else:
|
64 |
+
llm = None
|
65 |
return llm
|
66 |
|
67 |
|
|
|
70 |
temperature: float,
|
71 |
):
|
72 |
if model_name is None:
|
73 |
+
model = "gpt-3.5-turbo"
|
74 |
elif model_name == "rinna":
|
75 |
model = "rinna"
|
76 |
elif model_name == "GPT-3.5":
|
|
|
163 |
return result["result"], html
|
164 |
|
165 |
|
166 |
+
AVAILABLE_LLMS = ["GPT-3.5", "GPT-4"]
|
167 |
+
|
168 |
+
if RINNA_MODEL is not None:
|
169 |
+
AVAILABLE_LLMS.append("rinna")
|
170 |
+
|
171 |
nvdajp_book_qa = gr.Interface(
|
172 |
fn=main,
|
173 |
inputs=[
|
174 |
gr.Textbox(label="query"),
|
175 |
gr.Radio(["E5", "OpenAI"], label="Embedding", info="選択なしで「E5」を使用"),
|
176 |
+
gr.Radio(AVAILABLE_LLMS, label="Model", info="選択なしで「GPT-3.5」を使用"),
|
177 |
gr.Radio(
|
178 |
["All", "ja-book", "ja-nvda-user-guide", "en-nvda-user-guide"],
|
179 |
label="絞り込み",
|