File size: 7,826 Bytes
19b9c91
52f5630
19b9c91
476a48c
 
 
 
 
19b9c91
 
 
 
 
476a48c
52f5630
 
 
 
 
 
 
 
 
 
 
19b9c91
 
 
 
 
 
 
 
 
52f5630
 
 
19b9c91
 
 
52f5630
 
476a48c
 
 
 
 
 
 
 
19b9c91
476a48c
 
19b9c91
 
476a48c
 
 
19b9c91
476a48c
 
 
19b9c91
476a48c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0aeb372
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
476a48c
19b9c91
476a48c
 
19b9c91
476a48c
 
 
 
0aeb372
 
 
 
 
19b9c91
 
 
 
 
 
b999014
476a48c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
import os
from importlib.metadata import version
from inspect import currentframe, getframeinfo
from pathlib import Path

from decouple import config
from theflow.settings.default import *  # noqa

cur_frame = currentframe()
if cur_frame is None:
    raise ValueError("Cannot get the current frame.")
this_file = getframeinfo(cur_frame).filename
this_dir = Path(this_file).parent

# change this if your app use a different name
KH_PACKAGE_NAME = "kotaemon_app"

KH_APP_VERSION = None
try:
    # Caution: This might produce the wrong version
    # https://stackoverflow.com/a/59533071
    KH_APP_VERSION = version(KH_PACKAGE_NAME)
except Exception as e:
    print(f"Failed to get app version: {e}")

# App can be ran from anywhere and it's not trivial to decide where to store app data.
# So let's use the same directory as the flowsetting.py file.
KH_APP_DATA_DIR = this_dir / "ktem_app_data"
KH_APP_DATA_DIR.mkdir(parents=True, exist_ok=True)

# User data directory
KH_USER_DATA_DIR = KH_APP_DATA_DIR / "user_data"
KH_USER_DATA_DIR.mkdir(parents=True, exist_ok=True)

# doc directory
KH_DOC_DIR = this_dir / "docs"

# HF models can be big, let's store them in the app data directory so that it's easier
# for users to manage their storage.
# ref: https://huggingface.co/docs/huggingface_hub/en/guides/manage-cache
os.environ["HF_HOME"] = str(KH_APP_DATA_DIR / "huggingface")
os.environ["HF_HUB_CACHE"] = str(KH_APP_DATA_DIR / "huggingface")

COHERE_API_KEY = config("COHERE_API_KEY", default="")
KH_MODE = "dev"
KH_FEATURE_USER_MANAGEMENT = False
KH_FEATURE_USER_MANAGEMENT_ADMIN = str(
    config("KH_FEATURE_USER_MANAGEMENT_ADMIN", default="admin")
)
KH_FEATURE_USER_MANAGEMENT_PASSWORD = str(
    config("KH_FEATURE_USER_MANAGEMENT_PASSWORD", default="XsdMbe8zKP8KdeE@")
)
KH_ENABLE_ALEMBIC = False
KH_DATABASE = f"sqlite:///{KH_USER_DATA_DIR / 'sql.db'}"
KH_FILESTORAGE_PATH = str(KH_USER_DATA_DIR / "files")

KH_DOCSTORE = {
    "__type__": "kotaemon.storages.SimpleFileDocumentStore",
    "path": str(KH_USER_DATA_DIR / "docstore"),
}
KH_VECTORSTORE = {
    "__type__": "kotaemon.storages.ChromaVectorStore",
    "path": str(KH_USER_DATA_DIR / "vectorstore"),
}
KH_LLMS = {}
KH_EMBEDDINGS = {}

# populate options from config
if config("AZURE_OPENAI_API_KEY", default="") and config(
    "AZURE_OPENAI_ENDPOINT", default=""
):
    if config("AZURE_OPENAI_CHAT_DEPLOYMENT", default=""):
        KH_LLMS["azure"] = {
            "spec": {
                "__type__": "kotaemon.llms.AzureChatOpenAI",
                "temperature": 0,
                "azure_endpoint": config("AZURE_OPENAI_ENDPOINT", default=""),
                "api_key": config("AZURE_OPENAI_API_KEY", default=""),
                "api_version": config("OPENAI_API_VERSION", default="")
                or "2024-02-15-preview",
                "azure_deployment": config("AZURE_OPENAI_CHAT_DEPLOYMENT", default=""),
                "timeout": 20,
            },
            "default": False,
            "accuracy": 5,
            "cost": 5,
        }
    if config("AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT", default=""):
        KH_EMBEDDINGS["azure"] = {
            "spec": {
                "__type__": "kotaemon.embeddings.AzureOpenAIEmbeddings",
                "azure_endpoint": config("AZURE_OPENAI_ENDPOINT", default=""),
                "api_key": config("AZURE_OPENAI_API_KEY", default=""),
                "api_version": config("OPENAI_API_VERSION", default="")
                or "2024-02-15-preview",
                "azure_deployment": config(
                    "AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT", default=""
                ),
                "timeout": 10,
            },
            "default": False,
        }

if config("OPENAI_API_KEY", default=""):
    KH_LLMS["openai"] = {
        "spec": {
            "__type__": "kotaemon.llms.ChatOpenAI",
            "temperature": 0,
            "base_url": config("OPENAI_API_BASE", default="")
            or "https://api.openai.com/v1",
            "api_key": config("OPENAI_API_KEY", default=""),
            "model": config("OPENAI_CHAT_MODEL", default="") or "gpt-3.5-turbo",
            "timeout": 10,
        },
        "default": False,
    }
    if len(KH_EMBEDDINGS) < 1:
        KH_EMBEDDINGS["openai"] = {
            "spec": {
                "__type__": "kotaemon.embeddings.OpenAIEmbeddings",
                "base_url": config("OPENAI_API_BASE", default="")
                or "https://api.openai.com/v1",
                "api_key": config("OPENAI_API_KEY", default=""),
                "model": config(
                    "OPENAI_EMBEDDINGS_MODEL", default="text-embedding-ada-002"
                )
                or "text-embedding-ada-002",
                "timeout": 10,
            },
            "default": False,
        }

if config("LOCAL_MODEL", default=""):
    KH_LLMS["local"] = {
        "spec": {
            "__type__": "kotaemon.llms.EndpointChatLLM",
            "endpoint_url": "http://localhost:31415/v1/chat/completions",
        },
        "default": False,
        "cost": 0,
    }
    if len(KH_EMBEDDINGS) < 1:
        KH_EMBEDDINGS["local"] = {
            "spec": {
                "__type__": "kotaemon.embeddings.EndpointEmbeddings",
                "endpoint_url": "http://localhost:31415/v1/embeddings",
            },
            "default": False,
            "cost": 0,
        }

OPENROUTER_API_KEY = config("OPENROUTER_API_KEY", default="")
openrouter_models = {
    "Mistral 7B Instruct (free)": "mistralai/mistral-7b-instruct:free",
    "Toppy M 7B (free)": "undi95/toppy-m-7b:free",
    "Meta: LLama 3 8B Instruct (free)": "meta-llama/llama-3-8b-instruct:free",
    "OpenChat 3.5 (free)": "openchat/openchat-7b:free",
    "MythoMist 7B (free)": "gryphe/mythomist-7b:free",
    "Nous: Capybara 7B (free)": "nousresearch/nous-capybara-7b:free",
    "Google: Gemma 7B (free)": "google/gemma-7b-it:free",
    "Cinematika 7B (alpha) (free)": "openrouter/cinematika-7b:free",
    "Hugging Face: Zephyr 7B (free)": "huggingfaceh4/zephyr-7b-beta:free",
    "RWKV v5: Eagle 7B": "recursal/eagle-7b",
}
for name, model in openrouter_models.items():
    KH_LLMS[f"OpenRouter - {name}"] = {
        "spec": {
            "__type__": "kotaemon.llms.ChatOpenAI",
            "base_url": "https://openrouter.ai/api/v1/",
            "api_key": OPENROUTER_API_KEY,
            "model": model,
        },
        "cost": 0,
        "default": model == "mistralai/mistral-7b-instruct:free",
    }

if len(KH_EMBEDDINGS) < 1:
    KH_EMBEDDINGS["local-bge-base-en-v1.5"] = {
        "spec": {
            "__type__": "kotaemon.embeddings.FastEmbedEmbeddings",
            "model_name": "BAAI/bge-base-en-v1.5",
        },
        "default": True,
    }

KH_REASONINGS = [
    "ktem.reasoning.simple.FullQAPipeline",
    "ktem.reasoning.react.ReactAgentPipeline",
    "ktem.reasoning.rewoo.RewooAgentPipeline",
]
KH_VLM_ENDPOINT = "{0}/openai/deployments/{1}/chat/completions?api-version={2}".format(
    config("AZURE_OPENAI_ENDPOINT", default=""),
    config("OPENAI_VISION_DEPLOYMENT_NAME", default="gpt-4-vision"),
    config("OPENAI_API_VERSION", default=""),
)


SETTINGS_APP = {
    "lang": {
        "name": "Language",
        "value": "en",
        "choices": [("English", "en"), ("Japanese", "ja")],
        "component": "dropdown",
    }
}


SETTINGS_REASONING = {
    "use": {
        "name": "Reasoning options",
        "value": None,
        "choices": [],
        "component": "radio",
    },
    "lang": {
        "name": "Language",
        "value": "en",
        "choices": [("English", "en"), ("Japanese", "ja")],
        "component": "dropdown",
    },
}


KH_INDEX_TYPES = ["ktem.index.file.FileIndex"]
KH_INDICES = [
    {
        "name": "File",
        "config": {},
        "index_type": "ktem.index.file.FileIndex",
    },
]