From ebf90d5c8234b901e40539b97a132a0bece5cffd Mon Sep 17 00:00:00 2001
From: freeneuro <923237475@qq.com>
Date: Thu, 17 Mar 2022 00:42:26 +0800
Subject: [PATCH] Update readme.md

---
 src/carla_dataset/readme.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/carla_dataset/readme.md b/src/carla_dataset/readme.md
index 76f01c7..05c9740 100644
--- a/src/carla_dataset/readme.md
+++ b/src/carla_dataset/readme.md
@@ -8,4 +8,5 @@
 + test_label_new folder as the same to train_label_new folder
 ## Note that
 1. After you download the dataset and place them into each folder, you should create a yaml in data folder, see data/carala.yaml for example. 
-2. To run compute the adversarial loss, you need to get the label(annotation) for each image. In our implementation, for simplicity, we used the Faster RCNN to inference each rendered image(due to the character of rendered image, we can get relatively accuracy label) and get the corresponding label.
\ No newline at end of file
+2. To run compute the adversarial loss, you need to get the label(annotation) for each image. In our implementation, for simplicity, we used the Faster RCNN to inference each rendered image(due to the character of rendered image, we can get relatively accuracy label) and get the corresponding label.
+3. We provide the used train_label in [https://pan.baidu.com/s/18ZcQQgsr1DFbyUK4YVQ8WQ](https://pan.baidu.com/s/18ZcQQgsr1DFbyUK4YVQ8WQ) with the fetech code idrl