michaelfeil commited on
Commit
bde25c8
1 Parent(s): c15e519

Upload sentence-transformers/all-MiniLM-L6-v2 ctranslate2 weights

Browse files
Files changed (2) hide show
  1. README.md +8 -4
  2. model.bin +2 -2
README.md CHANGED
@@ -38,7 +38,7 @@ Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on
38
 
39
  quantized version of [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2)
40
  ```bash
41
- pip install hf-hub-ctranslate2>=2.12.0 ctranslate2>=3.16.0
42
  ```
43
 
44
  ```python
@@ -78,16 +78,20 @@ embeddings = model.encode(
78
  print(embeddings.shape, embeddings)
79
  scores = (embeddings @ embeddings.T) * 100
80
 
 
 
 
 
81
  ```
82
 
83
- Checkpoint compatible to [ctranslate2>=3.16.0](https://github.com/OpenNMT/CTranslate2)
84
  and [hf-hub-ctranslate2>=2.12.0](https://github.com/michaelfeil/hf-hub-ctranslate2)
85
  - `compute_type=int8_float16` for `device="cuda"`
86
  - `compute_type=int8` for `device="cpu"`
87
 
88
- Converted on 2023-06-19 using
89
  ```
90
- ct2-transformers-converter --model sentence-transformers/all-MiniLM-L6-v2 --output_dir ~/tmp-ct2fast-all-MiniLM-L6-v2 --force --copy_files config_sentence_transformers.json tokenizer.json modules.json README.md tokenizer_config.json sentence_bert_config.json data_config.json vocab.txt special_tokens_map.json .gitattributes --trust_remote_code
91
  ```
92
 
93
  # Licence and other remarks:
 
38
 
39
  quantized version of [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2)
40
  ```bash
41
+ pip install hf-hub-ctranslate2>=2.12.0 ctranslate2>=3.17.1
42
  ```
43
 
44
  ```python
 
78
  print(embeddings.shape, embeddings)
79
  scores = (embeddings @ embeddings.T) * 100
80
 
81
+ # Hint: you can also host this code via REST API and
82
+ # via github.com/michaelfeil/infinity
83
+
84
+
85
  ```
86
 
87
+ Checkpoint compatible to [ctranslate2>=3.17.1](https://github.com/OpenNMT/CTranslate2)
88
  and [hf-hub-ctranslate2>=2.12.0](https://github.com/michaelfeil/hf-hub-ctranslate2)
89
  - `compute_type=int8_float16` for `device="cuda"`
90
  - `compute_type=int8` for `device="cpu"`
91
 
92
+ Converted on 2023-10-13 using
93
  ```
94
+ LLama-2 -> removed <pad> token.
95
  ```
96
 
97
  # Licence and other remarks:
model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2abb237beb39bae980a7537a16a1fe5a0f0be2184be1d9f39f755b731a582adc
3
- size 90857292
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e02198a1a1480129f35fede1751d0406a43e5ea8e7abb618ac58285e974cd6e
3
+ size 45430860