diff --git a/api/apps/api_app.py b/api/apps/api_app.py
index f294272232bf7c1c462aea9432801268ee81860b..cc6f646d7bf4f2224a11f95bed86335e3b996c40 100644
--- a/api/apps/api_app.py
+++ b/api/apps/api_app.py
@@ -105,8 +105,8 @@ def stats():
res = {
"pv": [(o["dt"], o["pv"]) for o in objs],
"uv": [(o["dt"], o["uv"]) for o in objs],
- "speed": [(o["dt"], o["tokens"]/o["duration"]) for o in objs],
- "tokens": [(o["dt"], o["tokens"]/1000.) for o in objs],
+ "speed": [(o["dt"], float(o["tokens"])/float(o["duration"])) for o in objs],
+ "tokens": [(o["dt"], float(o["tokens"])/1000.) for o in objs],
"round": [(o["dt"], o["round"]) for o in objs],
"thumb_up": [(o["dt"], o["thumb_up"]) for o in objs]
}
@@ -115,8 +115,7 @@ def stats():
return server_error_response(e)
-@manager.route('/new_conversation', methods=['POST'])
-@validate_request("user_id")
+@manager.route('/new_conversation', methods=['GET'])
def set_conversation():
token = request.headers.get('Authorization').split()[1]
objs = APIToken.query(token=token)
@@ -131,7 +130,7 @@ def set_conversation():
conv = {
"id": get_uuid(),
"dialog_id": dia.id,
- "user_id": req["user_id"],
+ "user_id": request.args.get("user_id", ""),
"message": [{"role": "assistant", "content": dia.prompt_config["prologue"]}]
}
API4ConversationService.save(**conv)
diff --git a/api/db/db_models.py b/api/db/db_models.py
index 27ad80f99d6fe3aaa19881dcf1d68de871026f61..e6f2d287ef079ae5ccbdd1e216fc174a21b3a7db 100644
--- a/api/db/db_models.py
+++ b/api/db/db_models.py
@@ -629,7 +629,7 @@ class Document(DataBaseModel):
max_length=128,
null=False,
default="local",
- help_text="where dose this document from")
+ help_text="where dose this document come from")
type = CharField(max_length=32, null=False, help_text="file extension")
created_by = CharField(
max_length=32,
diff --git a/deepdoc/parser/pdf_parser.py b/deepdoc/parser/pdf_parser.py
index dfd3756d26e2926a796f88431ea1e494b7507095..6c3324544dd1f2e552cbbfb1a7c43b402f427adc 100644
--- a/deepdoc/parser/pdf_parser.py
+++ b/deepdoc/parser/pdf_parser.py
@@ -43,7 +43,9 @@ class HuParser:
model_dir, "updown_concat_xgb.model"))
except Exception as e:
model_dir = snapshot_download(
- repo_id="InfiniFlow/text_concat_xgb_v1.0")
+ repo_id="InfiniFlow/text_concat_xgb_v1.0",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/deepdoc"),
+ local_dir_use_symlinks=False)
self.updown_cnt_mdl.load_model(os.path.join(
model_dir, "updown_concat_xgb.model"))
diff --git a/deepdoc/vision/layout_recognizer.py b/deepdoc/vision/layout_recognizer.py
index 917ee6ed83a48a2cbf61a0f33ea3608a3c314718..58ddcdb622d6ce12b7e37c09f946252b7900d935 100644
--- a/deepdoc/vision/layout_recognizer.py
+++ b/deepdoc/vision/layout_recognizer.py
@@ -43,7 +43,9 @@ class LayoutRecognizer(Recognizer):
"rag/res/deepdoc")
super().__init__(self.labels, domain, model_dir)
except Exception as e:
- model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc")
+ model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/deepdoc"),
+ local_dir_use_symlinks=False)
super().__init__(self.labels, domain, model_dir)
self.garbage_layouts = ["footer", "header", "reference"]
diff --git a/deepdoc/vision/ocr.py b/deepdoc/vision/ocr.py
index b55024ed456bb367047c50c0035a7af8710aeca3..d602da06fefada82796b4a5f1a0b6ca9720b96e5 100644
--- a/deepdoc/vision/ocr.py
+++ b/deepdoc/vision/ocr.py
@@ -486,7 +486,9 @@ class OCR(object):
self.text_detector = TextDetector(model_dir)
self.text_recognizer = TextRecognizer(model_dir)
except Exception as e:
- model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc")
+ model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/deepdoc"),
+ local_dir_use_symlinks=False)
self.text_detector = TextDetector(model_dir)
self.text_recognizer = TextRecognizer(model_dir)
diff --git a/deepdoc/vision/recognizer.py b/deepdoc/vision/recognizer.py
index 67e096ef2def315e1b51a1891e741491cbee89d2..1ca7c4478ffd5129851a424724d1981cf128f13e 100644
--- a/deepdoc/vision/recognizer.py
+++ b/deepdoc/vision/recognizer.py
@@ -41,7 +41,9 @@ class Recognizer(object):
"rag/res/deepdoc")
model_file_path = os.path.join(model_dir, task_name + ".onnx")
if not os.path.exists(model_file_path):
- model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc")
+ model_dir = snapshot_download(repo_id="InfiniFlow/deepdoc",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/deepdoc"),
+ local_dir_use_symlinks=False)
model_file_path = os.path.join(model_dir, task_name + ".onnx")
else:
model_file_path = os.path.join(model_dir, task_name + ".onnx")
diff --git a/deepdoc/vision/table_structure_recognizer.py b/deepdoc/vision/table_structure_recognizer.py
index 6779137d87b39383835c1b5b55682b9f41f6e222..548eb62046481d4cf708c532f49b45b74fa7940b 100644
--- a/deepdoc/vision/table_structure_recognizer.py
+++ b/deepdoc/vision/table_structure_recognizer.py
@@ -39,7 +39,9 @@ class TableStructureRecognizer(Recognizer):
get_project_base_directory(),
"rag/res/deepdoc"))
except Exception as e:
- super().__init__(self.labels, "tsr", snapshot_download(repo_id="InfiniFlow/deepdoc"))
+ super().__init__(self.labels, "tsr", snapshot_download(repo_id="InfiniFlow/deepdoc",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/deepdoc"),
+ local_dir_use_symlinks=False))
def __call__(self, images, thr=0.2):
tbls = super().__call__(images, thr)
diff --git a/rag/llm/embedding_model.py b/rag/llm/embedding_model.py
index e6ff0b117835819d34106d9f3dcedfc55560df04..e6e18fbed8fcea2fece6cb21b197fb619844b4ac 100644
--- a/rag/llm/embedding_model.py
+++ b/rag/llm/embedding_model.py
@@ -14,6 +14,8 @@
# limitations under the License.
#
from typing import Optional
+
+from huggingface_hub import snapshot_download
from zhipuai import ZhipuAI
import os
from abc import ABC
@@ -35,7 +37,10 @@ try:
query_instruction_for_retrieval="为这个句ĺ生ćčˇ¨ç¤şä»Ąç”¨äşŽćŁ€ç´˘ç›¸ĺ…łć–‡ç« ďĽš",
use_fp16=torch.cuda.is_available())
except Exception as e:
- flag_model = FlagModel("BAAI/bge-large-zh-v1.5",
+ model_dir = snapshot_download(repo_id="BAAI/bge-large-zh-v1.5",
+ local_dir=os.path.join(get_project_base_directory(), "rag/res/bge-large-zh-v1.5"),
+ local_dir_use_symlinks=False)
+ flag_model = FlagModel(model_dir,
query_instruction_for_retrieval="为这个句ĺ生ćčˇ¨ç¤şä»Ąç”¨äşŽćŁ€ç´˘ç›¸ĺ…łć–‡ç« ďĽš",
use_fp16=torch.cuda.is_available())