Merge pull request #1046 from 545999961/master

upda reranker evaluate code

Author: 545999961

FlagEmbedding/llm_reranker/README.md

@@ -3,6 +3,7 @@
 - [Model List](#model-list)
 - [Usage](#usage)
 - [Fine-tuning](#fine-tune)
+- [Evaluate Script](#evaluate-script)
 - [Evaluation](#evaluation)
 - [Citation](#citation)
 
@@ -374,7 +375,7 @@ If you download reranker-v2-minicpm-layerwise, you can load it with the followin
 
 ```shell
 python evaluate.py \
---input_path ../LLARA/data/finetune/toy_finetune_data.jsonl \
+--input_path ./toy_finetune_data.jsonl \
 --metrics mrr recall ndcg map precision \
 --k_values 1 10 100
 ```