[TIPC] Smaller batch size to avoid OOM, remove url annotation (PaddlePaddle#1665)

Author: LutaoChu

test_tipc/configs/bisenetv2/train_infer_python.txt

@@ -6,7 +6,7 @@ Global.use_gpu:null|null
 Global.auto_cast:null
 --iters:lite_train_lite_infer=50|lite_train_whole_infer=50|whole_train_whole_infer=1000
 --save_dir:
---batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=8
+--batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=3
 Global.pretrained_model:null
 train_model_name:best_model/model.pdparams
 train_infer_img_dir:test_tipc/data/cityscapes/val.list

test_tipc/configs/ocrnet_hrnetw18/train_infer_python.txt

@@ -6,7 +6,7 @@ Global.use_gpu:null|null
 Global.auto_cast:null
 --iters:lite_train_lite_infer=50|lite_train_whole_infer=50|whole_train_whole_infer=1000
 --save_dir:
---batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=8
+--batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=3
 Global.pretrained_model:null
 train_model_name:best_model/model.pdparams
 train_infer_img_dir:test_tipc/data/cityscapes/val.list

test_tipc/configs/segformer_b0/train_infer_python.txt

@@ -6,7 +6,7 @@ Global.use_gpu:null|null
 Global.auto_cast:null
 --iters:lite_train_lite_infer=50|lite_train_whole_infer=50|whole_train_whole_infer=1000
 --save_dir:
---batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=8
+--batch_size:lite_train_lite_infer=2|lite_train_whole_infer=2|whole_train_whole_infer=3
 Global.pretrained_model:null
 train_model_name:best_model/model.pdparams
 train_infer_img_dir:test_tipc/data/cityscapes/val.list

test_tipc/docs/test_train_inference_python.md

@@ -2,8 +2,6 @@
 
 Linux端基础训练预测功能测试的主程序为`test_train_inference_python.sh`，可以测试基于Python的模型训练、评估、推理等基本功能，包括裁剪、量化、蒸馏。
 
-<!-- - Mac端基础训练预测功能测试参考[链接](./mac_test_train_inference_python.md)
-- Windows端基础训练预测功能测试参考[链接](./win_test_train_inference_python.md) -->
 
 ## 1. 测试结论汇总