remove 'tar' suffix and improve yolox-tiny from 31.7 to 32.8 (#329)

README.md

@@ -33,7 +33,7 @@ For more details, please refer to our [report on Arxiv](https://arxiv.org/abs/21
 |Model |size |mAP<sup>val<br>0.5:0.95 | Params<br>(M) |FLOPs<br>(G)| weights |
 | ------        |:---:  |  :---:       |:---:     |:---:  | :---: |
 |[YOLOX-Nano](./exps/default/nano.py) |416  |25.3  | 0.91 |1.08 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/EdcREey-krhLtdtSnxolxiUBjWMy6EFdiaO9bdOwZ5ygCQ?e=yQpdds)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_nano.pth) |
-|[YOLOX-Tiny](./exps/default/yolox_tiny.py) |416  |31.7 | 5.06 |6.45 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/EYtjNFPqvZBBrQ-VowLcSr4B6Z5TdTflUsr_gO2CwhC3bQ?e=SBTwXj)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_tiny.pth) |
+|[YOLOX-Tiny](./exps/default/yolox_tiny.py) |416  |32.8 | 5.06 |6.45 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/EbZuinX5X1dJmNy8nqSRegABWspKw3QpXxuO82YSoFN1oQ?e=Q7V7XE)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_tiny_32dot8.pth) |
 
 ## Quick Start
 
@@ -71,15 +71,15 @@ Step1. Download a pretrained model from the benchmark table.
 Step2. Use either -n or -f to specify your detector's config. For example:
 
 ```shell
-python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 or
 ```shell
-python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 Demo for video:
 ```shell
-python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth.tar --path /path/to/your/video --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth --path /path/to/your/video --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 
 
@@ -130,7 +130,7 @@ python tools/train.py -f exps/default/yolox-s.py -d 8 -b 64 --fp16 -o
 We support batch testing for fast evaluation:
 
 ```shell
-python tools/eval.py -n  yolox-s -c yolox_s.pth.tar -b 64 -d 8 --conf 0.001 [--fp16] [--fuse]
+python tools/eval.py -n  yolox-s -c yolox_s.pth -b 64 -d 8 --conf 0.001 [--fp16] [--fuse]
                          yolox-m
                          yolox-l
                          yolox-x
@@ -141,7 +141,7 @@ python tools/eval.py -n  yolox-s -c yolox_s.pth.tar -b 64 -d 8 --conf 0.001 [--f
 
 To reproduce speed test, we use the following command:
 ```shell
-python tools/eval.py -n  yolox-s -c yolox_s.pth.tar -b 1 -d 1 --conf 0.001 --fp16 --fuse
+python tools/eval.py -n  yolox-s -c yolox_s.pth -b 1 -d 1 --conf 0.001 --fp16 --fuse
                          yolox-m
                          yolox-l
                          yolox-x

demo/MegEngine/python/README.md

@@ -13,7 +13,7 @@ python3 -m pip install megengine -f https://megengine.org.cn/whl/mge.html
 ### Step2: convert checkpoint weights from torch's path file
 
 ```
-python3 convert_weights.py -w yolox_s.pth.tar -o yolox_s_mge.pkl
+python3 convert_weights.py -w yolox_s.pth -o yolox_s_mge.pkl
 ```
 
 ### Step3: run demo

demo/ONNXRuntime/README.md

@@ -6,7 +6,7 @@ This doc introduces how to convert your pytorch model into onnx, and how to run
 | Model | Parameters | GFLOPs | Test Size | mAP | Weights |
 |:------| :----: | :----: | :---: | :---: | :---: |
 |  YOLOX-Nano |  0.91M  | 1.08 | 416x416 | 25.3 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/EfAGwvevU-lNhW5OqFAyHbwBJdI_7EaKu5yU04fgF5BU7w?e=gvq4hf)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_nano.onnx) |
-|  YOLOX-Tiny | 5.06M     | 6.45 | 416x416 |31.7 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/EVigCszU1ilDn-MwLwHCF1ABsgTy06xFdVgZ04Yyo4lHVA?e=hVKiCw)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_tiny.onnx) |
+|  YOLOX-Tiny | 5.06M     | 6.45 | 416x416 |32.8 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/ET64VPoEV8FAm5YBiEj5JXwBVn_KYHM38iJQ_lpcK2slYw?e=uuJ7Ii)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_tiny_32dot8.onnx) |
 |  YOLOX-S | 9.0M | 26.8 | 640x640 |39.6 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/Ec0L1d1x2UtIpbfiahgxhtgBZVjb1NCXbotO8SCOdMqpQQ?e=siyIsK)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_s.onnx) |
 |  YOLOX-M | 25.3M | 73.8 | 640x640 |46.4 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/ERUKlQe-nlxBoTKPy1ynbxsBmAZ_h-VBEV-nnfPdzUIkZQ?e=hyQQtl)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_m.onnx) |
 |  YOLOX-L | 54.2M | 155.6 | 640x640 |50.0 | [onedrive](https://megvii-my.sharepoint.cn/:u:/g/personal/gezheng_megvii_com/ET5w926jCA5GlVfg9ixB4KEBiW0HYl7SzaHNRaRG9dYO_A?e=ISmCYX)/[github](https://github.com/Megvii-BaseDetection/storage/releases/download/0.0.1/yolox_l.onnx) |
@@ -24,7 +24,7 @@ Then, you can:
 
 1. Convert a standard YOLOX model by -n:
 ```shell
-python3 tools/export_onnx.py --output-name yolox_s.onnx -n yolox-s -c yolox_s.pth.tar
+python3 tools/export_onnx.py --output-name yolox_s.onnx -n yolox-s -c yolox_s.pth
 ```
 Notes:
 * -n: specify a model name. The model name must be one of the [yolox-s,m,l,x and yolox-nane, yolox-tiny, yolov3]
@@ -40,13 +40,13 @@ Notes:
 2. Convert a standard YOLOX model by -f. When using -f, the above command is equivalent to:
 
 ```shell
-python3 tools/export_onnx.py --output-name yolox_s.onnx -f exps/default/yolox_s.py -c yolox_s.pth.tar
+python3 tools/export_onnx.py --output-name yolox_s.onnx -f exps/default/yolox_s.py -c yolox_s.pth
 ```
 
 3. To convert your customized model, please use -f:
 
 ```shell
-python3 tools/export_onnx.py --output-name your_yolox.onnx -f exps/your_dir/your_yolox.py -c your_yolox.pth.tar
+python3 tools/export_onnx.py --output-name your_yolox.onnx -f exps/your_dir/your_yolox.py -c your_yolox.pth
 ```
 
 ### ONNXRuntime Demo

demo/TensorRT/python/README.md

@@ -16,7 +16,7 @@ YOLOX models can be easily conveted to TensorRT models using torch2trt
    ```
    For example:
    ```shell
-   python tools/trt.py -n yolox-s -c your_ckpt.pth.tar
+   python tools/trt.py -n yolox-s -c your_ckpt.pth
    ```
    <YOLOX_MODEL_NAME> can be: yolox-nano, yolox-tiny. yolox-s, yolox-m, yolox-l, yolox-x.
 
@@ -26,7 +26,7 @@ YOLOX models can be easily conveted to TensorRT models using torch2trt
    ```
    For example:
    ```shell
-   python tools/trt.py -f /path/to/your/yolox/exps/yolox_s.py -c your_ckpt.pth.tar
+   python tools/trt.py -f /path/to/your/yolox/exps/yolox_s.py -c your_ckpt.pth
    ```
    *yolox_s.py* can be any exp file modified by you.
 

docs/train_custom_data.md

@@ -73,13 +73,13 @@ python tools/train.py -f /path/to/your/Exp/file -d 8 -b 64 --fp16 -o -c /path/to
 
 or take the `YOLOX-S` VOC training for example:
 ```bash
-python tools/train.py -f exps/example/yolox_voc/yolox_voc_s.py -d 8 -b 64 --fp16 -o -c /path/to/yolox_s.pth.tar
+python tools/train.py -f exps/example/yolox_voc/yolox_voc_s.py -d 8 -b 64 --fp16 -o -c /path/to/yolox_s.pth
 ```
 
 ✧✧✧ For example:
 - If you download the [mini-coco128](https://drive.google.com/file/d/16N3u36ycNd70m23IM7vMuRQXejAJY9Fs/view?usp=sharing) and unzip it to the `datasets`, you can direct run the following training code.
     ```bash
-    python tools/train.py -f exps/example/custom/yolox_s.py -d 8 -b 64 --fp16 -o -c /path/to/yolox_s.pth.tar
+    python tools/train.py -f exps/example/custom/yolox_s.py -d 8 -b 64 --fp16 -o -c /path/to/yolox_s.pth
     ```
 
 (Don't worry for the different shape of detection head between the pretrained weights and your own model, we will handle it)

tools/demo.py

@@ -255,7 +255,7 @@ def main(exp, args):
 
     if not args.trt:
         if args.ckpt is None:
-            ckpt_file = os.path.join(file_name, "best_ckpt.pth.tar")
+            ckpt_file = os.path.join(file_name, "best_ckpt.pth")
         else:
             ckpt_file = args.ckpt
         logger.info("loading checkpoint")

tools/eval.py

@@ -147,7 +147,7 @@ def main(exp, args, num_gpu):
 
     if not args.speed and not args.trt:
         if args.ckpt is None:
-            ckpt_file = os.path.join(file_name, "best_ckpt.pth.tar")
+            ckpt_file = os.path.join(file_name, "best_ckpt.pth")
         else:
             ckpt_file = args.ckpt
         logger.info("loading checkpoint")

tools/export_onnx.py

@@ -63,7 +63,7 @@ def main():
     model = exp.get_model()
     if args.ckpt is None:
         file_name = os.path.join(exp.output_dir, args.experiment_name)
-        ckpt_file = os.path.join(file_name, "best_ckpt.pth.tar")
+        ckpt_file = os.path.join(file_name, "best_ckpt.pth")
     else:
         ckpt_file = args.ckpt
 

tools/trt.py

@@ -42,7 +42,7 @@ def main():
     file_name = os.path.join(exp.output_dir, args.experiment_name)
     os.makedirs(file_name, exist_ok=True)
     if args.ckpt is None:
-        ckpt_file = os.path.join(file_name, "best_ckpt.pth.tar")
+        ckpt_file = os.path.join(file_name, "best_ckpt.pth")
     else:
         ckpt_file = args.ckpt
 

yolox/core/trainer.py

@@ -269,7 +269,7 @@ class Trainer:
         if self.args.resume:
             logger.info("resume training")
             if self.args.ckpt is None:
-                ckpt_file = os.path.join(self.file_name, "latest" + "_ckpt.pth.tar")
+                ckpt_file = os.path.join(self.file_name, "latest" + "_ckpt.pth")
             else:
                 ckpt_file = self.args.ckpt
 

yolox/utils/checkpoint.py

@@ -37,8 +37,8 @@ def load_ckpt(model, ckpt):
 def save_checkpoint(state, is_best, save_dir, model_name=""):
     if not os.path.exists(save_dir):
         os.makedirs(save_dir)
-    filename = os.path.join(save_dir, model_name + "_ckpt.pth.tar")
+    filename = os.path.join(save_dir, model_name + "_ckpt.pth")
     torch.save(state, filename)
     if is_best:
-        best_filename = os.path.join(save_dir, "best_ckpt.pth.tar")
+        best_filename = os.path.join(save_dir, "best_ckpt.pth")
         shutil.copyfile(filename, best_filename)