pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 12:32:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix task cancling bug (#98)

Browse Source
This commit is contained in:
KevinHuSh
2024-03-05 16:33:47 +08:00
committed by GitHub
parent 07d76ea18d
commit 602038ac49
11 changed files with 24 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
rag/app/book.py
View File

@ -28,7 +28,7 @@ class Pdf(PdfParser):
from_page,
to_page,
callback)
callback("OCR finished")
callback(msg="OCR finished")
from timeit import default_timer as timer
start = timer()

4
rag/app/laws.py
View File

@ -57,7 +57,7 @@ class Pdf(PdfParser):
to_page,
callback
)
callback("OCR finished")
callback(msg="OCR finished")
from timeit import default_timer as timer
start = timer()
@ -135,6 +135,6 @@ def chunk(filename, binary=None, from_page=0, to_page=100000, lang="Chinese", ca
if __name__ == "__main__":
import sys
def dummy(a, b):
def dummy(prog=None, msg=""):
pass
chunk(sys.argv[1], callback=dummy)

2
rag/app/manual.py
View File

@ -22,7 +22,7 @@ class Pdf(PdfParser):
to_page,
callback
)
callback("OCR finished.")
callback(msg="OCR finished.")
from timeit import default_timer as timer
start = timer()

2
rag/app/naive.py
View File

@ -29,7 +29,7 @@ class Pdf(PdfParser):
to_page,
callback
)
callback("OCR finished")
callback(msg="OCR finished")
from timeit import default_timer as timer
start = timer()

2
rag/app/paper.py
View File

@ -36,7 +36,7 @@ class Pdf(PdfParser):
to_page,
callback
)
callback("OCR finished.")
callback(msg="OCR finished.")
from timeit import default_timer as timer
start = timer()

9
rag/nlp/search.py
View File

@ -305,8 +305,15 @@ class Dealer:
"similarity": sim[i],
"vector_similarity": vsim[i],
"term_similarity": tsim[i],
"vector": self.trans2floats(sres.field[id].get("q_%d_vec" % dim, "\t".join(["0"] * dim)))
"vector": self.trans2floats(sres.field[id].get("q_%d_vec" % dim, "\t".join(["0"] * dim))),
"positions": sres.field[id].get("position_int", "").split("\t")
}
if len(d["positions"]) % 5 == 0:
poss = []
for i in range(0, len(d["positions"]), 5):
poss.append([float(d["positions"][i]), float(d["positions"][i + 1]), float(d["positions"][i + 2]),
float(d["positions"][i + 3]), float(d["positions"][i + 4])])
d["positions"] = poss
ranks["chunks"].append(d)
if dnm not in ranks["doc_aggs"]:
ranks["doc_aggs"][dnm] = {"doc_id": did, "count": 0}

6
rag/svr/task_executor.py
View File

@ -25,6 +25,7 @@ import traceback
from functools import partial
from timeit import default_timer as timer
import numpy as np
from elasticsearch_dsl import Q
from api.db.services.task_service import TaskService
@ -177,10 +178,11 @@ def embedding(docs, mdl, parser_config={}, callback=None):
tts, c = mdl.encode(tts)
tk_count += c
cnts_ = []
cnts_ = np.array([])
for i in range(0, len(cnts), 32):
vts, c = mdl.encode(cnts[i: i+32])
cnts_.extend(vts)
if len(cnts_) == 0: cnts_ = vts
else: cnts_ = np.concatenate((cnts_, vts), axis=0)
tk_count += c
callback(msg="")
cnts = cnts_