diff --git a/configs/mot/bytetrack/_base_/ppyoloe_mot_reader_640x640.yml b/configs/mot/bytetrack/_base_/ppyoloe_mot_reader_640x640.yml
index c1e7ab8418956810ae6d2788cd4d67b9f2e17775..ef6342fd0e9249acf386b7795cb538b73a26f108 100644
--- a/configs/mot/bytetrack/_base_/ppyoloe_mot_reader_640x640.yml
+++ b/configs/mot/bytetrack/_base_/ppyoloe_mot_reader_640x640.yml
@@ -1,4 +1,8 @@
-worker_num: 8
+worker_num: 4
+eval_height: &eval_height 640
+eval_width: &eval_width 640
+eval_size: &eval_size [*eval_height, *eval_width]
+
 TrainReader:
   sample_transforms:
     - Decode: {}
@@ -20,17 +24,17 @@ TrainReader:
 EvalReader:
   sample_transforms:
     - Decode: {}
-    - Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
+    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
     - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
     - Permute: {}
   batch_size: 8
 
 TestReader:
   inputs_def:
-    image_shape: [3, 640, 640]
+    image_shape: [3, *eval_height, *eval_width]
   sample_transforms:
     - Decode: {}
-    - Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
+    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
     - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
     - Permute: {}
   batch_size: 1
@@ -40,17 +44,17 @@ TestReader:
 EvalMOTReader:
   sample_transforms:
     - Decode: {}
-    - Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
+    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
     - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
     - Permute: {}
   batch_size: 1
 
 TestMOTReader:
   inputs_def:
-    image_shape: [3, 640, 640]
+    image_shape: [3, *eval_height, *eval_width]
   sample_transforms:
     - Decode: {}
-    - Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
+    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
     - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
     - Permute: {}
   batch_size: 1
diff --git a/configs/mot/bytetrack/bytetrack_ppyoloe.yml b/configs/mot/bytetrack/bytetrack_ppyoloe.yml
index 08b7a00d89b79ad1bd1e2753738f22fcc66e657c..5e7ffe07f0f758c641596e90ee0da4c31085fd85 100644
--- a/configs/mot/bytetrack/bytetrack_ppyoloe.yml
+++ b/configs/mot/bytetrack/bytetrack_ppyoloe.yml
@@ -8,7 +8,7 @@ weights: output/bytetrack_ppyoloe/model_final
 log_iter: 20
 snapshot_epoch: 2
 
-metric: MOT # eval/infer mode
+metric: MOT # eval/infer mode, set 'COCO' can be training mode
 num_classes: 1
 
 architecture: ByteTrack
@@ -33,7 +33,6 @@ PPYOLOEHead:
   grid_cell_offset: 0.5
   static_assigner_epoch: -1 # 100
   use_varifocal_loss: True
-  eval_input_size: [640, 640]
   loss_weight: {class: 1.0, iou: 2.5, dfl: 0.5}
   static_assigner:
     name: ATSSAssigner
diff --git a/configs/mot/bytetrack/bytetrack_ppyoloe_pplcnet.yml b/configs/mot/bytetrack/bytetrack_ppyoloe_pplcnet.yml
index 98ea15f15299b9d550bc4de1a53fe203e7cd61fc..60f81165d5b324943a997dbc26fbe56f249f2ef6 100644
--- a/configs/mot/bytetrack/bytetrack_ppyoloe_pplcnet.yml
+++ b/configs/mot/bytetrack/bytetrack_ppyoloe_pplcnet.yml
@@ -33,7 +33,6 @@ PPYOLOEHead:
   grid_cell_offset: 0.5
   static_assigner_epoch: -1 # 100
   use_varifocal_loss: True
-  eval_input_size: [640, 640]
   loss_weight: {class: 1.0, iou: 2.5, dfl: 0.5}
   static_assigner:
     name: ATSSAssigner
diff --git a/configs/mot/bytetrack/detector/ppyoloe_crn_l_36e_640x640_mot17half.yml b/configs/mot/bytetrack/detector/ppyoloe_crn_l_36e_640x640_mot17half.yml
index 89654f059e603eda24002dfec844f450bd73e8ff..6c770e9bf85e953a30df43faf57c401518b7f6ad 100644
--- a/configs/mot/bytetrack/detector/ppyoloe_crn_l_36e_640x640_mot17half.yml
+++ b/configs/mot/bytetrack/detector/ppyoloe_crn_l_36e_640x640_mot17half.yml
@@ -7,6 +7,7 @@ weights: output/ppyoloe_crn_l_36e_640x640_mot17half/model_final
 log_iter: 20
 snapshot_epoch: 2
 
+
 # schedule configuration for fine-tuning
 epoch: 36
 LearningRate:
@@ -16,7 +17,7 @@ LearningRate:
       max_epochs: 43
     - !LinearWarmup
       start_factor: 0.001
-      steps: 100
+      epochs: 1
 
 OptimizerBuilder:
   optimizer:
@@ -26,9 +27,11 @@ OptimizerBuilder:
     factor: 0.0005
     type: L2
 
+
 TrainReader:
   batch_size: 8
 
+
 # detector configuration
 architecture: YOLOv3
 norm_type: sync_bn
@@ -63,7 +66,6 @@ PPYOLOEHead:
   grid_cell_offset: 0.5
   static_assigner_epoch: -1 # 100
   use_varifocal_loss: True
-  eval_input_size: [640, 640]
   loss_weight: {class: 1.0, iou: 2.5, dfl: 0.5}
   static_assigner:
     name: ATSSAssigner