Closed YangShuguang closed 6 years ago
您好，我想把单字类别扩展一下，把setting中的1000改掉后，想预加载模型，在train_image_classifer.py中把最后一层的名字给屏蔽掉：'checkpoint_exclude_scopes', ‘alexnet_v2/fc8’。发现仍会报错，似乎一直会预加载fc8层： InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [6855] rhs shape= [1001]，[[Node: save_1/Assign_44 = Assign[T=DT_FLOAT, _class=["loc:@alexnet_v2/fc8/biases"], use_locking=true 能给个思路吗？
预加载模型地址和保存模型的地址要分开，大坑
YangShuguang
commented
6 years ago YangShuguang
commented
6 years ago
您好，我想把单字类别扩展一下，把setting中的1000改掉后，想预加载模型，在train_image_classifer.py中把最后一层的名字给屏蔽掉：'checkpoint_exclude_scopes', ‘alexnet_v2/fc8’。发现仍会报错，似乎一直会预加载fc8层： InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [6855] rhs shape= [1001]，[[Node: save_1/Assign_44 = Assign[T=DT_FLOAT, _class=["loc:@alexnet_v2/fc8/biases"], use_locking=true 能给个思路吗？