djovak commited on
Commit
13a780c
1 Parent(s): d1a532e

update readme.md

Browse files
Files changed (1) hide show
  1. README.md +8 -1
README.md CHANGED
@@ -45,9 +45,16 @@ print(embeddings)
45
  ```
46
 
47
  ### Important usage notes
48
- - "ošišana ćirilica" (usage of c instead of ć, etc...) significantly deacreases search quality
49
  - The usage of uppercase letters for named entities can significantly improve search quality
50
 
 
 
 
 
 
 
 
51
 
52
  ## Evaluation
53
 
 
45
  ```
46
 
47
  ### Important usage notes
48
+ - "ošišana latinica" (usage of c instead of ć, etc...) significantly deacreases search quality
49
  - The usage of uppercase letters for named entities can significantly improve search quality
50
 
51
+ ## Training
52
+
53
+ - Embedić models are fine-tuned from multilingual-e5 models and they come in 3 sizes (small, base, large).
54
+
55
+ - Training is done on a single 4070ti super GPU
56
+
57
+ - 3-step training: distillation, training on (query, text) pairs and finally fine-tuning with triplets.
58
 
59
  ## Evaluation
60