Code

Embedding update#2791

Code

Open

pengjunfeng11 wants to merge 26 commits intoxorbitsai:mainxorbitsai/inference:mainfrom pengjunfeng11:embedding-updatepengjunfeng11/inference:embedding-updateCopy head branch name to clipboard

+1,444-112

Commits

Commits on Nov 10, 2024

• 支持使用bge-m3模型生成稀疏向量

  

  pengjunfeng11

  committed
  a2aedfb

  Copy full SHA for a2aedfb

Commits on Nov 11, 2024

• 新增embedding模型通过token id转换为字符的方法create_embedding

  

  pengjunfeng11

  committed
  46196d0

  Copy full SHA for 46196d0

• 调用create_embedding时根据参数判断是否调用flagEmbedding

  

  pengjunfeng11

  committed
  94b0cb7

  Copy full SHA for 94b0cb7

• sparse vector support

  

  pengjunfeng11

  committed
  fa8ae35

  Copy full SHA for fa8ae35

• Merge branch 'main' of https://github.com/pengjunfeng11/inference into main

  Show description for 37c2763
  

  pengjunfeng11

  committed
  37c2763

  Copy full SHA for 37c2763

• FEATURE: bge-m3 embedding model genarate sparse vector support

  

  pengjunfeng11

  committed
  da03958

  Copy full SHA for da03958

Commits on Nov 15, 2024

• Update core.py

  pengjunfeng11

  authored
  78735d0

  Copy full SHA for 78735d0

• Update core.py

  pengjunfeng11

  authored
  7e113bb

  Copy full SHA for 7e113bb

Commits on Nov 19, 2024

• FEAT: test from convert_ids_to_tokens method

  

  Junfeng Peng

  committed
  db7c849

  Copy full SHA for db7c849

• Merge branch 'main' of https://github.com/pengjunfeng11/inference into main

  Show description for 37a7692
  

  Junfeng Peng

  committed
  37a7692

  Copy full SHA for 37a7692

Commits on Nov 20, 2024

• FEAT: Remove reload in core.py, add test func for convert_ids_to_tokens

  

  Junfeng Peng

  committed
  0e7c6c4

  Copy full SHA for 0e7c6c4

• FEAT:Remove reload in core.py, add test func for convert_ids_to_tokens

  

  Junfeng Peng

  committed
  82ba2ec

  Copy full SHA for 82ba2ec

• FEAT:Remove reload in core.py, add test func for convert_ids_to_tokens

  

  Junfeng Peng

  committed
  bd5dd66

  Copy full SHA for bd5dd66

Commits on Nov 21, 2024

• FEAT:Remove reload in create_embedding method. Added test func for convert_ids_to_tokens. Added _encode_bgem3 method for bge-m3 model encoding

  

  Junfeng Peng

  committed
  cbff470

  Copy full SHA for cbff470

Commits on Nov 22, 2024

• FEAT:Remove reload in create_embedding method. Added test func for convert_ids_to_tokens. Added _encode_bgem3 method for bge-m3 model encoding

  

  Junfeng Peng

  committed
  9bcce28

  Copy full SHA for 9bcce28

Commits on Dec 26, 2024

• Update llm_family.py

  Show description for 5c1313b
  pengjunfeng11

  authored
  5c1313b

  Copy full SHA for 5c1313b

• Update llm_family.py

  Show description for 35cf7be
  pengjunfeng11

  authored
  35cf7be

  Copy full SHA for 35cf7be

• Update llm_family.py

  Show description for e597823
  pengjunfeng11

  authored
  e597823

  Copy full SHA for e597823

• Update llm_family.py

  Show description for 95953a5
  pengjunfeng11

  authored
  95953a5

  Copy full SHA for 95953a5

• Update llm_family.py

  Show description for 9b05e0e
  pengjunfeng11

  authored
  9b05e0e

  Copy full SHA for 9b05e0e

Commits on Jan 27, 2025

• demo

  

  Junfeng Peng

  committed
  3d294b8

  Copy full SHA for 3d294b8

Commits on Jan 28, 2025

• frist

  

  Junfeng Peng

  committed
  973bf90

  Copy full SHA for 973bf90

• merge branch 'main' of github.com:pengjunfeng11/inference into main

  

  Junfeng Peng

  committed
  46cc1a3

  Copy full SHA for 46cc1a3

• third

  

  Junfeng Peng

  committed
  1b8ee1e

  Copy full SHA for 1b8ee1e

• fourth

  

  Junfeng Peng

  committed
  e38f637

  Copy full SHA for e38f637

• fifth

  

  Junfeng Peng

  committed
  4d1bac0

  Copy full SHA for 4d1bac0