Update app.py
Browse files
app.py
CHANGED
@@ -2,18 +2,15 @@ import spaces
|
|
2 |
import subprocess
|
3 |
import gradio as gr
|
4 |
from huggingface_hub import hf_hub_download
|
5 |
-
from llama_cpp import Llama
|
6 |
-
from llama_cpp_agent import LlamaCppAgent
|
7 |
-
from llama_cpp_agent import MessagesFormatterType
|
8 |
-
from llama_cpp_agent.providers import LlamaCppPythonProvider
|
9 |
|
10 |
-
|
11 |
-
from llama_index.llms
|
12 |
-
from llama_index.llms.llama_cpp
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
|
|
17 |
|
18 |
subprocess.run('pip install llama-cpp-python==0.2.75 --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu124', shell=True)
|
19 |
subprocess.run('pip install llama-cpp-agent', shell=True)
|
@@ -29,6 +26,11 @@ def respond(
|
|
29 |
temperature,
|
30 |
top_p,
|
31 |
):
|
|
|
|
|
|
|
|
|
|
|
32 |
llama_model = Llama(r"models/mistral-7b-instruct-v0.2.Q6_K.gguf", n_batch=1024, n_threads=0, n_gpu_layers=33, n_ctx=8192, verbose=False)
|
33 |
|
34 |
provider = LlamaCppPythonProvider(llama_model)
|
|
|
2 |
import subprocess
|
3 |
import gradio as gr
|
4 |
from huggingface_hub import hf_hub_download
|
|
|
|
|
|
|
|
|
5 |
|
6 |
+
|
7 |
+
# from llama_index.core.llms import ChatMessage, MessageRole
|
8 |
+
# from llama_index.llms.llama_cpp import LlamaCPP
|
9 |
+
# from llama_index.llms.llama_cpp.llama_utils import (
|
10 |
+
# messages_to_prompt,
|
11 |
+
# completion_to_prompt,
|
12 |
+
# )
|
13 |
+
# from llama_index.core.memory import ChatMemoryBuffer
|
14 |
|
15 |
subprocess.run('pip install llama-cpp-python==0.2.75 --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu124', shell=True)
|
16 |
subprocess.run('pip install llama-cpp-agent', shell=True)
|
|
|
26 |
temperature,
|
27 |
top_p,
|
28 |
):
|
29 |
+
from llama_cpp import Llama
|
30 |
+
from llama_cpp_agent import LlamaCppAgent
|
31 |
+
from llama_cpp_agent import MessagesFormatterType
|
32 |
+
from llama_cpp_agent.providers import LlamaCppPythonProvider
|
33 |
+
|
34 |
llama_model = Llama(r"models/mistral-7b-instruct-v0.2.Q6_K.gguf", n_batch=1024, n_threads=0, n_gpu_layers=33, n_ctx=8192, verbose=False)
|
35 |
|
36 |
provider = LlamaCppPythonProvider(llama_model)
|