-
Notifications
You must be signed in to change notification settings - Fork 0
/
root_config.py
executable file
·134 lines (104 loc) · 5.53 KB
/
root_config.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
import os
class RootConfig:
# global variables
logSaver = None
tempModelCatch = []
tempPipeLineKnowledgeCatch = []
# ---------------------------------- The following configurations need to be modified -----------------------------------
# 1. Absolute path of the project(default: the path where the project is located).
# root_path = os.path.dirname(os.path.abspath(__file__))
# 2. If you start with Docker, the path is this.
root_path = "/app/KMatrix_v2"
print("--------------root_path-------------\n",root_path)
if not root_path.endswith("/"):
root_path = root_path + "/"
dir_init_config_path = root_path + "dir_init_config/"
dir_task_reult_data_path = root_path + "dir_task_reult_data/"
datasetUploadDirPath = root_path + "dir_dataset_upload/"
knowledgeUploadDirPath = root_path + "dir_knowledge_upload/"
# GPU used in the project
CUDA_VISIBLE_DEVICES = "1"
# Server PORT
SERVER_PORT = "8020"
# Default ES connection address
ES_HOST = "http://127.0.0.1:9200"
ES_USERNAME = "username"
ES_PASSWORD = "password"
# OPENAI api_key
openai_api_key = "sk-xxx"
openai_model_version = "gpt-4o"
# google_search_key You can go to https://serpapi.com/ to get it
google_search_key = "xxx"
# Proxy (used for openai api calls)
HTTP_PROXY = "http://127.0.0.1:xxxx"
HTTPS_PROXY = "http://127.0.0.1:xxxx"
# ----------------------------------The following configurations do not need to be modified -----------------------------------
model_path = root_path
# self_rag model
selfRAG_model_path = model_path + "dir_model/generator/selfrag_llama2_7b"
if not os.path.exists(selfRAG_model_path):
print("selfRAG Does not exist locally, requires online loading...")
selfRAG_model_path = "selfrag/selfrag_llama2_7b"
# llama2 model
llama2_model_path = model_path + "dir_model/generator/Llama-2-7b-chat-hf"
if not os.path.exists(llama2_model_path):
print("llama2 Does not exist locally, requires online loading...")
llama2_model_path = "meta-llama/Llama-2-7b-chat-hf"
# baichuan2 model
baichuan2_model_path = model_path + "dir_model/generator/Baichuan2-13B-Chat"
if not os.path.exists(baichuan2_model_path):
print("baichuan2 Does not exist locally, requires online loading...")
baichuan2_model_path = "baichuan-inc/Baichuan2-13B-Chat"
NED_model_path = model_path + "dir_model/generator/NED_model"
if not os.path.exists(NED_model_path):
print("NED Does not exist locally, requires online loading...")
NED_model_path = "wikipedia_model_with_numbers"
WikiSP_model_path = model_path + "dir_model/generator/WikiSP_sparql_model"
if not os.path.exists(WikiSP_model_path):
print("WikiSP Does not exist locally, requires online loading...")
WikiSP_model_path = "stanford-oval/llama-7b-wikiwebquestions-qald7"
# retriever_model_path
# BGE model
BGE_model_path = model_path + "dir_model/retriever/BGE/BGE-Reproduce"
if not os.path.exists(BGE_model_path):
print("BGE Does not exist locally, requires online loading...")
BGE_model_path = "BAAI/bge-large-en-v1.5"
BGEm3_model_path = "BAAI/bge-m3"
# contriever model
contriever_model_path = model_path + "dir_model/retriever/contriever/contriever_msmarco_model"
if not os.path.exists(contriever_model_path):
print("contriever Does not exist locally, requires online loading...")
contriever_model_path = "nthakur/contriever-base-msmarco"
# DPR model
DPR_model_path = model_path + "dir_model/retriever/DPR/facebook-dpr-question_encoder-multiset-base"
if not os.path.exists(DPR_model_path):
print("DPR Does not exist locally, requires online loading...")
DPR_model_path = "sentence-transformers/facebook-dpr-question_encoder-multiset-base"
# BERT model
BERT_model_path = model_path + "dir_model/retriever/BERT/Bert-base"
if not os.path.exists(BERT_model_path):
print("BERT Does not exist locally, requires online loading...")
BERT_model_path = "google-bert/bert-base-uncased"
# E5 model
E5_model_path = model_path + "dir_model/retriever/E5/e5-mistral-7b-instruct"
if not os.path.exists(E5_model_path):
print("E5 Does not exist locally, requires online loading...")
E5_model_path = "intfloat/e5-mistral-7b-instruct"
# Qwen2.5-14BE5 model
Qwen25_14B_model_path = model_path + "dir_model/retriever/Qwen2.5-14B-Instruct"
if not os.path.exists(Qwen25_14B_model_path):
print("Qwen25_14B Does not exist locally, requires online loading...")
Qwen25_14B_model_path = "/mnt/publiccache/huggingface/Qwen2.5-14B-Instruct"
# # query - model
wikidata_base_model_path = "LLama_2_7b_hf"
wikidata_peft_model_path = "llama-2-7b-sparql-8bit"
scienceqa_model_path = "princeton-nlp/sup-simcse-roberta-large"
# verbalizer model
verbalizer_model_path = model_path + "dir_model/verlizer/t5-large_T-F_ID-T/val_avg_bleu=54.5600-step_count=3.ckpt"
if not os.path.exists(verbalizer_model_path):
print("verbalizer Does not exist locally, requires online loading...")
verbalizer_model_path = ""
T5_model_path = model_path + "dir_model/verlizer/t5-large"
if not os.path.exists(T5_model_path):
print("T5 Does not exist locally, requires online loading...")
T5_model_path = "google-t5/t5-large"