yolov5离线量化报错

[luoshiyong]

----------- Running Arguments -----------
arch: YOLOv5
dataset_dir: /mnt/data01/luoshiyong/data/liugang
model_dir: /mnt/data01/luoshiyong/code/PaddleSlim-release-2.5/yolov5out/best.onnx
skip_tensors: None
train_anno_path: annotations/instances1_train.json
train_image_dir: images/train
val_anno_path: annotations/instances1_val.json
val_image_dir: images/val

loading annotations into memory...
Done (t=0.22s)
creating index...
index created!
I0626 11:09:01.155823 541758 interpretercore.cc:237] New Executor is Running.
2024-06-26 11:09:01,300-INFO: Loaded model from: /mnt/data01/luoshiyong/code/PaddleSlim-release-2.5/yolov5out/best_infer
Wed Jun 26 11:09:01-INFO: Load model and set data loader ...
Wed Jun 26 11:09:01-WARNING: The old way to load inference model is deprecated. Please specify path_prefix. model path: /mnt/data01/luoshiyong/code/PaddleSlim-release-2.5/yolov5out/best_infer/model.pdmodel, params path: /mnt/data01/luoshiyong/code/PaddleSlim-release-2.5/yolov5out/best_infer/model.pdiparams
W0626 11:09:01.377684 541758 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version: 12.2, Runtime API Version: 11.6
W0626 11:09:01.419098 541758 gpu_resources.cc:149] device: 0, cuDNN Version: 8.4.
Wed Jun 26 11:09:05-INFO: Collect quantized variable names ...
Sampling stage, Run batch:| | 0/32
Traceback (most recent call last):
File "post_quant.py", line 98, in
main()
File "post_quant.py", line 72, in main
quant_post_static(
File "/home/cisdi/anaconda3/envs/luoshiyong/lib/python3.8/site-packages/paddleslim-Stopping.at.filesystem.boundary.GIT_DISCOVERY_ACROSS_FILESYSTEM.not.set.-py3.8.egg/paddleslim/quant/quanter.py", line 669, in quant_post_static
File "/home/cisdi/anaconda3/envs/luoshiyong/lib/python3.8/site-packages/paddle/static/quantization/post_training_quantization.py", line 482, in quantize
self._executor.run(
File "/home/cisdi/anaconda3/envs/luoshiyong/lib/python3.8/site-packages/paddle/fluid/executor.py", line 1392, in run
res = self._run_impl(
File "/home/cisdi/anaconda3/envs/luoshiyong/lib/python3.8/site-packages/paddle/fluid/executor.py", line 1618, in _run_impl
ret = new_exe.run(
File "/home/cisdi/anaconda3/envs/luoshiyong/lib/python3.8/site-packages/paddle/fluid/executor.py", line 654, in run
tensors = self._new_exe.run(
ValueError: (InvalidArgument) The 'shape' in ReshapeOp is invalid. The input tensor X'size must be equal to the capacity of 'shape'. But received X's shape = [1, 512, 1, 1], X's size = 512, 'shape' is [2, 512], the capacity of 'shape' is 1024.
[Hint: Expected capacity == in_size, but received capacity:1024 != in_size:512.] (at ../paddle/fluid/operators/reshape_op.cc:234)
[operator < reshape2 > error]

[minghaoBD]

你好，这里的错误是指 reshape op的输入X和目标shape无法兼容。可以尝试将目标shape改成 [-1, 512]