RT-DETR官方代码已发布 #8114

lyuwenyu · 2023-04-20T01:48:10Z

论文: DETRs Beat YOLOs on Real-time Object Detection
代码:

简介

RT-DETR是第一个实时端到端目标检测器。具体而言，我们设计了一个高效的混合编码器，通过解耦尺度内交互和跨尺度融合来高效处理多尺度特征，并提出了IoU感知的查询选择机制，以优化解码器查询的初始化。此外，RT-DETR支持通过使用不同的解码器层来灵活调整推理速度，而不需要重新训练，这有助于实时目标检测器的实际应用。RT-DETR-L在COCO val2017上实现了53.0%的AP，在T4 GPU上实现了114FPS，RT-DETR-X实现了54.8%的AP和74FPS，在速度和精度方面都优于相同规模的所有YOLO检测器。RT-DETR-R50实现了53.1%的AP和108FPS，RT-DETR-R101实现了54.3%的AP和74FPS，在精度上超过了全部使用相同骨干网络的DETR检测器。
若要了解更多细节，请参考我们的论文paper.

lyuwenyu · 2023-04-21T02:41:44Z

如何训练自己的数据集呢

https://github.com/PaddlePaddle/PaddleDetection/blob/develop/docs/tutorials/data/PrepareDetDataSet.md

LeoCho1learning · 2023-04-21T03:12:04Z

请问有没有对比过INT8下的速度表现呢

lyuwenyu · 2023-04-21T03:45:00Z

请问有没有对比过INT8下的速度表现呢

已经在做了，( 如果用INT8 RT-DETR可能优势会更明显 ( 总的时间缩短但是带后处理的模型后处理时间是稳定的

1117wwh · 2023-04-21T07:33:52Z

不能eval,只能训练
(pytorch) C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop>python -u tools/eval.py -c configs/rtdetr/rtdetr_r50vd_6x_coco.yml -o weights=output/rtdetr_r50vd_6x_coco/8.pdparams
Warning: Unable to use JDE/FairMOT/ByteTrack, please install lap, for example: pip install lap, see https://github.com/gatagat/lap
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0421 15:32:24.823109 15072 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version: 11.5, Runtime API Version: 11.2
W0421 15:32:24.826112 15072 gpu_resources.cc:91] device: 0, cuDNN Version: 8.3.
loading annotations into memory...
Done (t=0.03s)
creating index...
index created!
[04/21 15:32:26] ppdet.data.source.coco INFO: Load [600 samples valid, 0 samples invalid] in file dataset/coco\annotations/instances_train2017.json.
[04/21 15:32:27] ppdet.utils.checkpoint INFO: Finish loading model weights: output/rtdetr_r50vd_6x_coco/8.pdparams
Traceback (most recent call last):
File "tools/eval.py", line 208, in
main()
File "tools/eval.py", line 204, in main
run(FLAGS, cfg)
File "tools/eval.py", line 159, in run
trainer.evaluate()
File "C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop\ppdet\engine\trainer.py", line 672, in evaluate
self._eval_with_loader(self.loader)
File "C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop\ppdet\engine\trainer.py", line 641, in _eval_with_loader
metric.update(data, outs)
File "C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop\ppdet\metrics\metrics.py", line 114, in update
infer_results = get_infer_results(
File "C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop\ppdet\metrics\coco_utils.py", line 52, in get_infer_results
infer_res['bbox'] = get_det_res(
File "C:\Users\lenovo\Desktop\study\PaddleDetection-develop\PaddleDetection-develop\ppdet\metrics\json_results.py", line 30, in get_det_res
category_id = label_to_cat_id_map[int(num_id)]
KeyError: 1

lyuwenyu · 2023-04-21T08:20:38Z

训练的自己的数据吧, 应该是val.json里的类别不全， category_id = label_to_cat_id_map[int(num_id)]换成category_id = int(num_id)试一下，，或者参考这个issue #4777

itsrealkk · 2023-04-21T08:31:00Z

请问如何对转出的onnx和trt进行推理呢谢谢

TsingWei · 2023-04-21T08:36:49Z

看代码发现HybridEncoder里连MSDeformableAttention都去掉了？好厉害。

lyuwenyu · 2023-04-21T09:07:27Z

请问如何对转出的onnx和trt进行推理呢谢谢

在readme里有具体操作的步骤

jingjingli066 · 2023-04-22T11:26:27Z

请问 File "D:\Code\LJJ\rtdetr\PaddleDetection\ppdet\core\workspace.py", line 229, in create
raise ValueError("The module {} is not registered".format(name))
ValueError: The module RTDETRTransformer is not registered
这是什么原因呢？

shengs1113 · 2023-04-23T04:40:32Z

预训练模型的下载路径，文档的描述和配置文件里的默认值不一致。
hgnet_v2用2个预训练模型都提示不匹配；提示信息in model is not matched with pretrained weights, and its will be trained from scratch
r50vd可以进入训练，在log出第1个iter信息后卡死。
机器是windows环境，rtx3090GPU

wm901115nwpu · 2023-04-23T14:16:29Z

PP-YOLOE+_l 的 mAP 是 52点多，yolov5_l是67.3，论文的图中为什么mAP那么低？

lyuwenyu · 2023-04-24T02:02:59Z

请问 File "D:\Code\LJJ\rtdetr\PaddleDetection\ppdet\core\workspace.py", line 229, in create raise ValueError("The module {} is not registered".format(name)) ValueError: The module RTDETRTransformer is not registered 这是什么原因呢？

pull最新的代码试一下，我们这边测试没有出现你说的这个情况

hhxdestiny · 2023-04-24T11:50:38Z

请问有添加到FastDeploy的计划么？

hhxdestiny · 2023-04-25T01:08:27Z

PP-YOLOE+_l 的 mAP 是 52点多，yolov5_l是67.3，论文的图中为什么mAP那么低？

看看论文，你看窜列了，你看的是红色这一列

wm901115nwpu · 2023-04-25T02:29:17Z

好的

lyuwenyu · 2023-04-25T02:55:25Z

请问有添加到FastDeploy的计划么？

已经在做了 @hhxdestiny

wacdev · 2023-04-25T03:51:07Z

能给一个怎么使用onnx模型的演示代码吗？

直接用官方教程，会报错

❯ python tools/infer.py -c configs/rtdetr/rtdetr_r101vd_6x_coco.yml \
 -o weights=/root/art/paddle/model/rtdetr_r101vd_6x_coco.pdparams \
 --infer_img=/root/art/paddle/test/1.jpg

[04/25 16:34:22] ppdet.utils.checkpoint INFO: Finish loading model weights: /root/art/paddle/model/rtdetr_r101vd_6x_coco.pdparams
[04/25 16:34:22] ppdet.data.source.category WARNING: anno_file 'dataset/coco/annotations/instances_val2017.json' is None or not set or not exist, please recheck TrainDataset/EvalDataset/Test
Dataset.anno_path, otherwise the default categories will be used by metric_type.
[04/25 16:34:22] ppdet.data.source.category WARNING: metric_type: COCO, load default categories of COCO.
  0%|                                                                                                                                                                   | 0/1 [00:00<?, ?it/s]
[04/25 16:34:22] reader WARNING: fail to map sample transform [Decode_13aee1] with error: 'bytes' object has no attribute 'shape' and stack:
Traceback (most recent call last):
  File "/root/art/paddle/PaddleDetection/ppdet/data/reader.py", line 59, in __call__
    data = f(data)
  File "/root/art/paddle/PaddleDetection/ppdet/data/transform/operators.py", line 105, in __call__
    sample = self.apply(sample, context)
  File "/root/art/paddle/PaddleDetection/ppdet/data/transform/operators.py", line 138, in apply
    sample['h'] = im.shape[0]
AttributeError: 'bytes' object has no attribute 'shape'

Exception in thread Thread-1 (_thread_loop):
Traceback (most recent call last):
  File "/root/.local/share/rtx/installs/python/3.10.11/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/root/.local/share/rtx/installs/python/3.10.11/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/root/art/paddle/.direnv/python-3.10.11/lib/python3.10/site-packages/paddle/fluid/dataloader/dataloader_iter.py", line 247, in _thread_loop
    batch = self._dataset_fetcher.fetch(
  File "/root/art/paddle/.direnv/python-3.10.11/lib/python3.10/site-packages/paddle/fluid/dataloader/fetcher.py", line 127, in fetch
    data.append(self.dataset[idx])
  File "/root/art/paddle/PaddleDetection/ppdet/data/source/dataset.py", line 102, in __getitem__
    return self.transform(roidb)
  File "/root/art/paddle/PaddleDetection/ppdet/data/reader.py", line 65, in __call__
    raise e
  File "/root/art/paddle/PaddleDetection/ppdet/data/reader.py", line 59, in __call__
    data = f(data)
  File "/root/art/paddle/PaddleDetection/ppdet/data/transform/operators.py", line 105, in __call__
    sample = self.apply(sample, context)
  File "/root/art/paddle/PaddleDetection/ppdet/data/transform/operators.py", line 138, in apply
    sample['h'] = im.shape[0]
AttributeError: 'bytes' object has no attribute 'shape'

用官方的 rtdetr_r101vd_6x_coco.onnx 和这篇文章中的代码 https://zhuanlan.zhihu.com/p/622940435

运行还是报错

Traceback (most recent call last):

File "/Users/z/git/paddle/./test.py", line 48, in

output = session.run(None, {'image': img})

File "/Users/z/.local/share/rtx/installs/python/3.10.11/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 196, in run

raise ValueError("Model requires {} inputs. Input Feed contains {}".format(num_required_inputs, num_inputs))

ValueError: Model requires 3 inputs. Input Feed contains 1

改成 https://zhuanlan.zhihu.com/p/622940435 中的的模型才能运行

Originlightwkp · 2023-04-25T07:01:18Z

我可以正常eval，但在训练中遇到这样一个问题，rtx2080 8g

/home/sc/.local/lib/python3.7/site-packages/matplotlib-3.4.0rc1-py3.7-linux-x86_64.egg/matplotlib/init.py:152: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
if LooseVersion(module.version) < minver:
/home/sc/development/anaconda3/envs/rtdetr/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
other = LooseVersion(other)
Warning: Unable to use JDE/FairMOT/ByteTrack, please install lap, for example: pip install lap, see https://github.com/gatagat/lap
Warning: Unable to use numba in PP-Tracking, please install numba, for example(python3.7): pip install numba==0.56.4
Warning: Unable to use numba in PP-Tracking, please install numba, for example(python3.7): pip install numba==0.56.4
Warning: Unable to use MOT metric, please install motmetrics, for example: pip install motmetrics, see https://github.com/longcw/py-motmetrics
Warning: Unable to use MCMOT metric, please install motmetrics, for example: pip install motmetrics, see https://github.com/longcw/py-motmetrics
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
loading annotations into memory...
Done (t=9.65s)
creating index...
index created!
[04/25 14:31:57] ppdet.data.source.coco WARNING: Found an invalid bbox in annotations: im_id: 200365, area: 0.0 x1: 296.65, y1: 388.33, x2: 297.67999999999995, y2: 388.33.
[04/25 14:32:08] ppdet.data.source.coco WARNING: Found an invalid bbox in annotations: im_id: 550395, area: 0.0 x1: 9.98, y1: 188.56, x2: 15.52, y2: 188.56.
[04/25 14:32:09] ppdet.data.source.coco INFO: Load [117266 samples valid, 1021 samples invalid] in file /media/sc/mSc/wkp/2017/annotations/instances_train2017.json.
W0425 14:32:10.895720 2548 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.5, Driver API Version: 11.4, Runtime API Version: 10.2
W0425 14:32:10.896754 2548 gpu_resources.cc:91] device: 0, cuDNN Version: 8.7.
[04/25 14:32:13] ppdet.utils.checkpoint INFO: ['fc.bias', 'fc.weight', 'last_conv.weight'] in pretrained weight is not used in the model, and its will not be loaded
[04/25 14:32:13] ppdet.utils.checkpoint INFO: Finish loading model weights: /home/sc/.cache/paddle/weights/PPHGNetV2_X_ssld_pretrained.pdparams

C++ Traceback (most recent call last):

0 paddle::pybind::ThrowExceptionToPython(std::__exception_ptr::exception_ptr)

Error Message Summary:

FatalError: Process abort signal is detected by the operating system.
[TimeInfo: *** Aborted at 1682404354 (unix time) try "date -d @1682404354" if you are using GNU date ***]
[SignalInfo: *** SIGABRT (@0x3e8000009f4) received by PID 2548 (TID 0x7f57b424f740) from PID 2548 ***]

/home/sc/code/rt-dert/train.sh: 行 2: 2548 已放弃 python PaddleDetection-develop/tools/train.py -c PaddleDetection-develop/confrtdetr/rtdetr_hgnetv2_x_6x_coco.yml --eval

sdreamforchen · 2023-05-04T07:57:34Z

请问有没有对比过INT8下的速度表现呢

已经在做了，( 如果用INT8 RT-DETR可能优势会更明显 ( 总的时间缩短但是带后处理的模型后处理时间是稳定的

您好，反馈一个我们某家国产芯片编译器出现的情况。麻烦大佬也能综合一下，考虑下我们部署环境的局限性，十分感谢。
出现了部分算子不支持的情况：Shape、Cast、Where、range、Topk、Tile、GatherND、GridSample、Mod，还有Slice算子，目前编译器仅支持通道维度上的切分。

eorroot · 2023-05-06T10:53:02Z

您好，有计划发布pytorch版本吗

zhiqwang · 2023-05-06T13:14:09Z

隔壁 ultralytics 已经在做 pytorch 版本了

leechenggg · 2023-05-06T15:21:54Z

用RTDETR进行SAHI切图拼图评估时报错

项目“RTDETR+SAHI切图”共享链接(有效期三天)：https://aistudio.baidu.com/studio/project/partial/verify/6092017/b32e0f8533b2482aa1675a99aa50aeec

/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/init.py:107: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
from collections import MutableMapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/rcsetup.py:20: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
from collections import Iterable, Mapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/colors.py:53: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
from collections import Sized
W0505 10:40:56.459722 895 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2
W0505 10:40:56.464571 895 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2.
loading annotations into memory...
Done (t=0.40s)
creating index...
index created!
[05/05 10:41:09] ppdet.data.source.coco INFO: 548 samples and slice to 2842 sub_samples in file /home/aistudio/MyDataset/val.json
[05/05 10:41:11] ppdet.utils.checkpoint INFO: Finish loading model weights: output/rtdetr_r101vd_6x_coco/best_model.pdparams
Traceback (most recent call last):
File "tools/eval.py", line 208, in
main()
File "tools/eval.py", line 204, in main
run(FLAGS, cfg)
File "tools/eval.py", line 157, in run
match_metric=FLAGS.match_metric)
File "/home/aistudio/PaddleDetection/ppdet/engine/trainer.py", line 764, in evaluate_slice
match_metric)
File "/home/aistudio/PaddleDetection/ppdet/engine/trainer.py", line 708, in _eval_with_loader_slice
outs['bbox'][:, 2:4] = outs['bbox'][:, 2:4] + shift_amount
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dygraph/math_op_patch.py", line 304, in impl
return math_op(self, other_var, -1)
ValueError: (InvalidArgument) Broadcast dimension mismatch. Operands could not be broadcast together with the shape of X = [1200, 2] and the shape of Y = [4, 2]. Received [1200] in X is not equal to [4] in Y at i:0.
[Hint: Expected x_dims_array[i] == y_dims_array[i] || x_dims_array[i] <= 1 || y_dims_array[i] <= 1 == true, but received x_dims_array[i] == y_dims_array[i] || x_dims_array[i] <= 1 || y_dims_array[i] <= 1:0 != true:1.] (at /paddle/paddle/phi/kernels/funcs/common_shape.h:84)

sdreamforchen · 2023-05-08T03:00:15Z

请问有没有对比过INT8下的速度表现呢

已经在做了，( 如果用INT8 RT-DETR可能优势会更明显 ( 总的时间缩短但是带后处理的模型后处理时间是稳定的

大佬，能否考虑尽可能的全量化，而不是根据敏感度分析，做部分量化。谢谢谢谢

universea · 2023-05-08T05:11:53Z

什么时候可以支持fastdeploy呢？

ly510244856 · 2023-05-08T14:09:20Z

我的输入数据是两张灰度图像，请问官方有相应的加载方法吗？或者有提供解决思路吗？

lyuwenyu · 2023-05-11T08:42:31Z

什么时候可以支持fastdeploy呢？

现在就已经支持的

lyuwenyu · 2023-05-11T08:43:05Z

我的输入数据是两张灰度图像，请问官方有相应的加载方法吗？或者有提供解决思路吗？

在backbone那个位置把输入的channel 3改成 1

lyuwenyu · 2023-05-11T08:43:45Z

请问有没有对比过INT8下的速度表现呢

已经在做了，( 如果用INT8 RT-DETR可能优势会更明显 ( 总的时间缩短但是带后处理的模型后处理时间是稳定的

大佬，能否考虑尽可能的全量化，而不是根据敏感度分析，做部分量化。谢谢谢谢

https://github.com/PaddlePaddle/PaddleDetection/tree/develop/deploy/auto_compression/configs

cqlp1314 · 2023-05-12T03:32:06Z

大佬，rt-detr结合swin或者convnext的话相比dino有优势吗？如果有的话可以我可以尝试结合一下

HuLu65 · 2023-05-13T10:03:25Z

怎么调整模型的保存间隔。默认是每次训练完都会保存一次。

cqlp1314 · 2023-05-16T02:14:19Z

使用--slice_infer时出现KeyError: 'st_pix',报错信息如下：
Traceback (most recent call last): File "tools/eval.py", line 209, in <module> main() File "tools/eval.py", line 205, in main run(FLAGS, cfg) File "tools/eval.py", line 158, in run match_metric=FLAGS.match_metric) File "/personal/cuichaoqi/bisai/PaddleDetection/ppdet/engine/trainer.py", line 764, in evaluate_slice match_metric) File "/personal/cuichaoqi/bisai/PaddleDetection/ppdet/engine/trainer.py", line 707, in _eval_with_loader_slice shift_amount = data['st_pix'] KeyError: 'st_pix'

ShawnNew · 2023-05-18T09:56:34Z

请问如何对转出的onnx和trt进行推理呢谢谢

在readme里有具体操作的步骤

@lyuwenyu 你好大佬，我在模型转onnx的时候强需求opset_version为11，修改之后paddle2onnx有如下信息

[Paddle2ONNX] Start to parse PaddlePaddle model...
[Paddle2ONNX] Model file path: ./output_inference/rtdetr_r101vd_6x_coco/model.pdmodel
[Paddle2ONNX] Paramters file path: ./output_inference/rtdetr_r101vd_6x_coco/model.pdiparams
[Paddle2ONNX] Start to parsing Paddle model...
[Paddle2ONNX] [grid_sampler: grid_sample_0.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_1.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_2.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_3.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_4.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_5.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_6.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_7.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_8.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_9.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_10.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_11.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_12.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_13.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_14.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_15.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_16.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] [grid_sampler: grid_sample_17.tmp_0] Requires the minimal opset version of 16.
[Paddle2ONNX] Due to the operator: grid_sampler, requires opset_version >= 16.
[Paddle2ONNX] Opset version will change to 16 from 11
[Paddle2ONNX] Use opset_version = 16 for ONNX export.
[Paddle2ONNX] PaddlePaddle model is exported as ONNX format now.

我可以如何绕过这个算子从而生成opset_version为11的模型？

MINGtoMING · 2023-05-22T08:46:27Z

@lyuwenyu develop版的deploy/auto_compression/configs/rtdetr_reader.yml文件的TrainDataset的COCODataSet设置重复了

dddlli · 2023-06-14T04:43:26Z

AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16,使用rtdetr_focalnet_L_384_3x_coco会报上述错误

jiluojiluo · 2023-06-15T06:43:30Z

AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16,使用rtdetr_focalnet_L_384_3x_coco会报上述错误
我也出现相同的错误。
AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16

sdreamforchen · 2023-06-15T10:04:53Z

1 datareader的TrainDataset未改；
2 量化训练还是无涨点，达不到官网的精度。上次测试也是这样
叨扰！

guoqsGary · 2023-06-18T09:59:47Z

TypeError: build_2d_sincos_position_embedding() takes from 2 to 4 positional arguments but 5 were given
请问为什么会出现这个问题呢？

whale1008 · 2023-06-20T03:07:31Z

请问一下，论文中说可以灵活使用不同的解码层而不需要重新训练，这个是怎么去实现/操作呢

TsingWei · 2023-06-20T04:47:21Z

请问一下，论文中说可以灵活使用不同的解码层而不需要重新训练，这个是怎么去实现/操作呢

@whale1008 我的看法是，因为训练的时候解码器的每一层都是在干同样的事情：输入box和feat，输出delta box，也就是对box进行refine。而且每一层的loss权重还是一样的，也就是说每一层的‘重要性’都一样。所以在推理时，可以使用全部的6层，也可以使用其中几层，甚至能只用第一层重复推理6次。

whale1008 · 2023-06-20T04:53:20Z

好的，感谢，推理时需要修改哪些配置参数呢

paipaipaidaxing · 2023-06-27T10:48:57Z

训练rtdetr_focalnet_L_384_3x_coco报错：
AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16

changzipeng · 2023-08-03T09:31:14Z

想问下论文里面的CSF和CCFM有什么区别呀

qianlian-mozi · 2023-08-19T07:36:45Z

佬，现在有pytorch版本的RT-DETR代码发布了么，我找了圈没找到

changzipeng · 2023-08-19T07:38:26Z

我倒是从网上搜到了一个rtdetr pytorch版本但是还没咋看应该不是官方发布的版本 | | ***@***.*** | | ***@***.*** |

…

---- 回复的原邮件 ---- | 发件人 | ***@***.***> | | 日期 | 2023年08月19日 15:36 | | 收件人 | ***@***.***> | | 抄送至 | ***@***.***>***@***.***> | | 主题 | Re: [PaddlePaddle/PaddleDetection] RT-DETR官方代码已发布 (Issue #8114) | 佬，现在有pytorch版本的RT-DETR代码发布了么，我找了圈没找到 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

qianlian-mozi · 2023-08-19T07:47:17Z

感谢回复，虽然好像看不清佬你邮件给我得信息，不过我重新找了找，找到了RT-DETR/models/rtdetr_l.py at main · Jinhui-Luan/RT-DETR (github.com)；Jinhui-Luan 大佬的pytorch复现版本，准备细看

…

------------------ 原始邮件 ------------------ 发件人: "PaddlePaddle/PaddleDetection" ***@***.***>; 发送时间: 2023年8月19日(星期六) 下午3:38 ***@***.***>; ***@***.******@***.***>; 主题: Re: [PaddlePaddle/PaddleDetection] RT-DETR官方代码已发布 (Issue #8114) 我倒是从网上搜到了一个rtdetr pytorch版本但是还没咋看应该不是官方发布的版本 | | ***@***.*** | | ***@***.*** |

---- 回复的原邮件 ---- | 发件人 | ***@***.***> | | 日期 | 2023年08月19日 15:36 | | 收件人 | ***@***.***> | | 抄送至 | ***@***.***>***@***.***> | | 主题 | Re: [PaddlePaddle/PaddleDetection] RT-DETR官方代码已发布 (Issue #8114) | 佬，现在有pytorch版本的RT-DETR代码发布了么，我找了圈没找到 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

OrionZJ · 2023-08-26T12:10:35Z

AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16,使用rtdetr_focalnet_L_384_3x_coco会报上述错误
我也出现相同的错误。
AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16

一样的错误，坐等修复

vscv · 2023-09-05T09:14:27Z

AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16,使用rtdetr_focalnet_L_384_3x_coco会报上述错误
我也出现相同的错误。
AssertionError: Variable dtype not match, Variable [ conv2d_0.w_0 ] need tensor with dtype paddle.float32 but load tensor with dtype paddle.float16

一样的错误，坐等修复

等大佬修復，謝謝。

#8605

m00nLi · 2023-09-20T06:47:58Z

实测精度太差了，训练自己的数据收敛都是个问题

l1uj1awe103 · 2023-11-30T12:59:11Z

Halo，RT-DETR有更佳性能的主干权重么？RT-DETR有蒸馏之类的计划么？

changzipeng · 2023-11-30T13:22:42Z

RT-DETR-X我看性能是最好的，backbone是hgnetv2，但是我之前修改模型结构，发现resnet101和hgnetv2基本差不多，可能hgnetv2会好一点，蒸馏之类的计划不太了解。

…

---- 回复的原邮件 ---- | 发件人 | ***@***.***> | | 发送日期 | 2023年11月30日 20:59 | | 收件人 | ***@***.***> | | 抄送人 | ***@***.***> , ***@***.***> | | 主题 | Re: [PaddlePaddle/PaddleDetection] RT-DETR官方代码已发布 (Issue #8114) | Halo，RT-DETR有更佳性能的主干权重么？RT-DETR有蒸馏之类的计划么？ — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

l1uj1awe103 · 2023-12-12T15:07:10Z

halo, 再请教下，RTDETR的静态图训练是不是有问题？

lyuwenyu added the feature request New feature or request label Apr 20, 2023

lyuwenyu pinned this issue Apr 20, 2023

lyuwenyu changed the title ~~RT-DETR代码已发布~~ **RT-DETR代码已发布** Apr 20, 2023

lyuwenyu changed the title ~~**RT-DETR代码已发布**~~ RT-DETR官方代码已发布 Apr 20, 2023

nemonameless mentioned this issue Apr 21, 2023

🚀 Roadmap of PaddleYOLO 🚀 PaddlePaddle/PaddleYOLO#44

Closed

4 tasks

ShawnNew mentioned this issue May 19, 2023

grid_sample算子导致paddle2onnx强制使用opset_version 16 PaddlePaddle/Paddle2ONNX#1102

Closed

lyuwenyu unpinned this issue Jun 28, 2023

paddle-bot bot assigned lyuwenyu Mar 8, 2024

RT-DETR官方代码已发布 #8114

RT-DETR官方代码已发布 #8114

Comments

lyuwenyu commented Apr 20, 2023 • edited

简介

lyuwenyu commented Apr 21, 2023

LeoCho1learning commented Apr 21, 2023

lyuwenyu commented Apr 21, 2023 • edited

1117wwh commented Apr 21, 2023

lyuwenyu commented Apr 21, 2023

itsrealkk commented Apr 21, 2023

TsingWei commented Apr 21, 2023

lyuwenyu commented Apr 21, 2023

jingjingli066 commented Apr 22, 2023

shengs1113 commented Apr 23, 2023 • edited

wm901115nwpu commented Apr 23, 2023

lyuwenyu commented Apr 24, 2023

hhxdestiny commented Apr 24, 2023

hhxdestiny commented Apr 25, 2023

wm901115nwpu commented Apr 25, 2023

lyuwenyu commented Apr 25, 2023

wacdev commented Apr 25, 2023 • edited

Originlightwkp commented Apr 25, 2023

C++ Traceback (most recent call last):

Error Message Summary:

sdreamforchen commented May 4, 2023

eorroot commented May 6, 2023

zhiqwang commented May 6, 2023

leechenggg commented May 6, 2023 • edited

sdreamforchen commented May 8, 2023

universea commented May 8, 2023

ly510244856 commented May 8, 2023

lyuwenyu commented May 11, 2023

lyuwenyu commented May 11, 2023

lyuwenyu commented May 11, 2023

cqlp1314 commented May 12, 2023

HuLu65 commented May 13, 2023

cqlp1314 commented May 16, 2023

ShawnNew commented May 18, 2023 • edited

MINGtoMING commented May 22, 2023

dddlli commented Jun 14, 2023

jiluojiluo commented Jun 15, 2023

sdreamforchen commented Jun 15, 2023

guoqsGary commented Jun 18, 2023

whale1008 commented Jun 20, 2023

TsingWei commented Jun 20, 2023

whale1008 commented Jun 20, 2023

paipaipaidaxing commented Jun 27, 2023

changzipeng commented Aug 3, 2023

qianlian-mozi commented Aug 19, 2023

changzipeng commented Aug 19, 2023 via email

qianlian-mozi commented Aug 19, 2023 via email

OrionZJ commented Aug 26, 2023

vscv commented Sep 5, 2023 • edited

m00nLi commented Sep 20, 2023

l1uj1awe103 commented Nov 30, 2023

changzipeng commented Nov 30, 2023 via email

l1uj1awe103 commented Dec 12, 2023

lyuwenyu commented Apr 20, 2023 •

edited

lyuwenyu commented Apr 21, 2023 •

edited

shengs1113 commented Apr 23, 2023 •

edited

wacdev commented Apr 25, 2023 •

edited

leechenggg commented May 6, 2023 •

edited

ShawnNew commented May 18, 2023 •

edited

vscv commented Sep 5, 2023 •

edited