pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Feature/feat1017 (#2872)

Browse Source 
### What problem does this PR solve?

1. fix: mid map show error in knowledge graph, juse because
```@antv/g6```version changed
2. feat: concurrent threads configuration support in graph extractor
3. fix: used tokens update failed for tenant
4. feat: timeout configuration support for llm
5. fix: regex error in graph extractor
6. feat: qwen rerank(```gte-rerank```) support
7. fix: timeout deal in knowledge graph index process. Now chat by
stream output, also, it is configuratable.
8. feat: ```qwen-long``` model configuration

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)

---------

Co-authored-by: chongchuanbing <chongchuanbing@gmail.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
This commit is contained in:
chongchuanbing
2024-10-21 12:11:08 +08:00
committed by GitHub
parent 4bdf3fd48e
commit ac26d09a59
8 changed files with 95 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
api/db/services/llm_service.py
View File

@ -167,11 +167,13 @@ class TenantLLMService(CommonService):
else:
assert False, "LLM type error"
llm_name = mdlnm.split("@")[0] if "@" in mdlnm else mdlnm
num = 0
try:
for u in cls.query(tenant_id=tenant_id, llm_name=mdlnm):
for u in cls.query(tenant_id=tenant_id, llm_name=llm_name):
num += cls.model.update(used_tokens=u.used_tokens + used_tokens)\
.where(cls.model.tenant_id == tenant_id, cls.model.llm_name == mdlnm)\
.where(cls.model.tenant_id == tenant_id, cls.model.llm_name == llm_name)\
.execute()
except Exception as e:
pass
@ -207,7 +209,7 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens):
database_logger.error(
"Can't update token usage for {}/EMBEDDING".format(self.tenant_id))
"Can't update token usage for {}/EMBEDDING used_tokens: {}".format(self.tenant_id, used_tokens))
return emd, used_tokens
def encode_queries(self, query: str):
@ -215,7 +217,7 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens):
database_logger.error(
"Can't update token usage for {}/EMBEDDING".format(self.tenant_id))
"Can't update token usage for {}/EMBEDDING used_tokens: {}".format(self.tenant_id, used_tokens))
return emd, used_tokens
def similarity(self, query: str, texts: list):
@ -223,7 +225,7 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens):
database_logger.error(
"Can't update token usage for {}/RERANK".format(self.tenant_id))
"Can't update token usage for {}/RERANK used_tokens: {}".format(self.tenant_id, used_tokens))
return sim, used_tokens
def describe(self, image, max_tokens=300):
@ -231,7 +233,7 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens):
database_logger.error(
"Can't update token usage for {}/IMAGE2TEXT".format(self.tenant_id))
"Can't update token usage for {}/IMAGE2TEXT used_tokens: {}".format(self.tenant_id, used_tokens))
return txt
def transcription(self, audio):
@ -239,7 +241,7 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens):
database_logger.error(
"Can't update token usage for {}/SEQUENCE2TXT".format(self.tenant_id))
"Can't update token usage for {}/SEQUENCE2TXT used_tokens: {}".format(self.tenant_id, used_tokens))
return txt
def tts(self, text):
@ -254,10 +256,10 @@ class LLMBundle(object):
def chat(self, system, history, gen_conf):
txt, used_tokens = self.mdl.chat(system, history, gen_conf)
if not TenantLLMService.increase_usage(
if isinstance(txt, int) and not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, used_tokens, self.llm_name):
database_logger.error(
"Can't update token usage for {}/CHAT".format(self.tenant_id))
"Can't update token usage for {}/CHAT llm_name: {}, used_tokens: {}".format(self.tenant_id, self.llm_name, used_tokens))
return txt
def chat_streamly(self, system, history, gen_conf):
@ -266,6 +268,6 @@ class LLMBundle(object):
if not TenantLLMService.increase_usage(
self.tenant_id, self.llm_type, txt, self.llm_name):
database_logger.error(
"Can't update token usage for {}/CHAT".format(self.tenant_id))
"Can't update token usage for {}/CHAT llm_name: {}, content: {}".format(self.tenant_id, self.llm_name, txt))
return
yield txt