yolov5prune icon indicating copy to clipboard operation
yolov5prune copied to clipboard

v7.0版本能用吗?

Open Doris-miao opened this issue 1 year ago • 4 comments

我用的是v5的7.0版本,稀疏训练的时候会报错,求解答 报错如下:Traceback (most recent call last): File "C:\Users\Admin\Desktop\jianzhi\yolov5prune-6.0\train_sparity.py", line 674, in main(opt) File "C:\Users\Admin\Desktop\jianzhi\yolov5prune-6.0\train_sparity.py", line 571, in main train(opt.hyp, opt, device, callbacks) File "C:\Users\Admin\Desktop\jianzhi\yolov5prune-6.0\train_sparity.py", line 117, in train ckpt = torch.load(weights, map_location=device) # load checkpoint File "C:\Users\Admin.conda\envs\yolov5-7.0\lib\site-packages\torch\serialization.py", line 607, in load return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args) File "C:\Users\Admin.conda\envs\yolov5-7.0\lib\site-packages\torch\serialization.py", line 882, in _load result = unpickler.load() File "C:\Users\Admin.conda\envs\yolov5-7.0\lib\site-packages\torch\serialization.py", line 875, in find_class return super().find_class(mod_name, name) AttributeError: Can't get attribute 'DetectionModel' on <module 'models.yolo' from 'C:\Users\Admin\Desktop\jianzhi\yolov5prune-6.0\models\yolo.py'>

Doris-miao avatar Sep 19 '23 03:09 Doris-miao

不能,要自己移植改一下捏

peiki99 avatar Nov 08 '23 02:11 peiki99

不能,要自己移植改一下捏 请问怎么移植,你有源代码吗?我现在需要实现v5 7.0版本的剪枝操作。

pppppsb avatar Nov 21 '23 08:11 pppppsb

不能,要自己移植改一下捏 请问怎么移植,你有源代码吗?我现在需要实现v5 7.0版本的剪枝操作。

在稀疏训练的时候可以直接把那一小块 加到 backward那里 然后 参数里面要剪枝的部分剪掉 用新的参数结构训练就行了 在v8上实测有效

Polaris-F avatar Dec 06 '23 14:12 Polaris-F

在v8上实测有效

你好,可以求指教怎么在yolov8上进行剪枝吗?有偿~谢谢~

wusaisa avatar Mar 05 '24 11:03 wusaisa