docs(README): 更新项目依赖并调整训练参数

- 更新了项目的依赖列表，增加了 yaml、ftfy、regex、tensorboard 等新依赖
- 调整了训练脚本 run_irra.sh 中的 batch_size 参数，从 64 修改为 16
- 更新了数据集根目录路径

README.md

@@ -20,10 +20,17 @@ The goal of this work is to enhance global text-to-image person retrieval perfor
 ### Requirements
 we use single RTX3090 24G GPU for training and evaluation. 
 ```
-pytorch 1.9.0
-torchvision 0.10.0
+pytorch==1.9.0
+torchvision==0.10.0
 prettytable
 easydict
+yaml
+ftfy
+regex
+tensorboard
+numpy==1.26.4
+tqdm
+setuptools==59.5.0
 ```
 
 ### Prepare Datasets

run_irra.sh

@@ -5,9 +5,9 @@ CUDA_VISIBLE_DEVICES=0 \
 python train.py \
 --name iira \
 --img_aug \
---batch_size 64 \
+--batch_size 16 \
 --MLM \
 --dataset_name $DATASET_NAME \
 --loss_names 'sdm+mlm+id' \
---root_dir '/mnt/vos-s9gjtkm2/reid/dataset/cross_reid' \
+--root_dir '/home/linkslinks/dataset' \
 --num_epoch 60