feat: update readme for deployment

README.md

@@ -17,7 +17,7 @@ YOLOv6 is composed of the following methods:
 ## Coming soon
 
 - [ ] YOLOv6 m/l/x model.
-- [ ] Deployment for OPENVINO/MNN/TNN/NCNN...
+- [ ] Deployment for MNN/TNN/NCNN/CoreML...
 
 
 ## Quick Start
@@ -73,13 +73,8 @@ python tools/eval.py --data data/coco.yaml  --batch 32 --weights yolov6s.pt --ta
 
 ### Deployment
 
-Export as ONNX Format
-
-```shell
-python deploy/export_onnx.py --weights yolov6s.pt --device 0
-                                       yolov6n.pt
-```
-
+*  [ONNX](./deploy/ONNX)
+*  [OpenVINO](./deploy/OpenVINO)
 
 ### Tutorials
 

deploy/ONNX/README.md

@@ -0,0 +1,14 @@
+## Export ONNX Model
+
+### Check requirements
+```shell
+pip install onnx>=1.10.0                             
+```
+
+### Export script
+```shell
+python deploy/ONNX/export_onnx.py --weights yolov6s.pt --img 640 --batch 1
+
+```
+
+### Download

deploy/export_onnx.py → deploy/ONNX/export_onnx.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python3
 # -*- coding:utf-8 -*-
-# https://github.com/ultralytics/yolov5/blob/master/export.py
 import argparse
 import time
 import sys
@@ -58,16 +57,6 @@
 
     y = model(img)  # dry run
 
-    # TorchScript export
-    try:
-        LOGGER.info('\nStarting to export TorchScript...')
-        export_file = args.weights.replace('.pt', '.torchscript.pt')  # filename
-        ts = torch.jit.trace(model, img)
-        ts.save(export_file)
-        LOGGER.info(f'TorchScript export success, saved as {export_file}')
-    except Exception as e:
-        LOGGER.info(f'TorchScript export failure: {e}')
-
     # ONNX export
     try:
         LOGGER.info('\nStarting to export ONNX...')

deploy/export_openvino.py → deploy/OpenVINO/export_openvino.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python3
 # -*- coding:utf-8 -*-
-# https://github.com/ultralytics/yolov5/blob/master/export.py
 import argparse
 import time
 import sys

docs/Test_speed.md

@@ -31,7 +31,7 @@ To  get inference speed with TensorRT in FP16 mode on T4, you can follow the ste
 First, export pytorch model as onnx format using the  following command:
 
 ```shell
-python deploy/export_onnx.py --weights yolov6n.pt --device 0 --batch [1 or 32] 
+python deploy/ONNX/export_onnx.py --weights yolov6n.pt --device 0 --batch [1 or 32] 
 ```
 
 Second,  generate an inference  trt engine and test speed using `trtexec`:

docs/Train_custom_data.md

@@ -125,6 +125,6 @@ python tools/infer.py --weights output_dir/name/weights/best_ckpt.pt --source im
 Export as ONNX Format
 
 ```shell
-python deploy/export_onnx.py --weights output_dir/name/weights/best_ckpt.pt --device 0
+python deploy/ONNX/export_onnx.py --weights output_dir/name/weights/best_ckpt.pt --device 0
 ```