Safetensors
Korean
new
reranker
korean
custom_code
sigridjineth commited on
Commit
0b791d5
1 Parent(s): 01a1eb9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -0
README.md CHANGED
@@ -79,6 +79,29 @@ The [AutoRAG Benchmark](https://github.com/Marker-Inc-Korea/AutoRAG-example-kore
79
  | 3 | 0.0813 | 0.3596 | 0.7193 | 0.2398 | 0.7061 | 0.7061 | 0.7096 | False |
80
  | 5 | 0.0824 | 0.3596 | 0.7193 | 0.2398 | 0.7061 | 0.7061 | 0.7096 | False |
81
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
82
  ## Usage (transformers>=4.36.0)
83
 
84
  ```python
 
79
  | 3 | 0.0813 | 0.3596 | 0.7193 | 0.2398 | 0.7061 | 0.7061 | 0.7096 | False |
80
  | 5 | 0.0824 | 0.3596 | 0.7193 | 0.2398 | 0.7061 | 0.7061 | 0.7096 | False |
81
 
82
+ ```
83
+ Evaluation Results (k=1,3,5,10):
84
+ Accuracy@1: 0.8070
85
+ F1@1: 0.8070
86
+ Recall@1: 0.8070
87
+ Precision@1: 0.8070
88
+ Accuracy@3: 0.9211
89
+ F1@3: 0.4605
90
+ Recall@3: 0.9211
91
+ Precision@3: 0.3070
92
+ Accuracy@5: 0.9474
93
+ F1@5: 0.3158
94
+ Recall@5: 0.9474
95
+ Precision@5: 0.1895
96
+ Accuracy@10: 0.9737
97
+ F1@10: 0.1770
98
+ Recall@10: 0.9737
99
+ Precision@10: 0.0974
100
+
101
+ Total inference time (all queries): 142.64 sec
102
+ Average inference time (per query): 1.2512 sec
103
+ ```
104
+
105
  ## Usage (transformers>=4.36.0)
106
 
107
  ```python