You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
如文中所提出的Fisher-diag方式来估计Hessian矩阵,需要计算每一层pre-activation的梯度。但在实际代码运行时,save_grad_data中的cur_grad = get_grad(cali_data[i * batch_size:(i + 1) * batch_size])在执行到第二个batch的时候会报错Trying to backward through the graph a second time,第一个batch的数据并不会报错。不知道作者是否遇到过类似的情况?
The text was updated successfully, but these errors were encountered:
如文中所提出的Fisher-diag方式来估计Hessian矩阵,需要计算每一层pre-activation的梯度。但在实际代码运行时,save_grad_data中的cur_grad = get_grad(cali_data[i * batch_size:(i + 1) * batch_size])在执行到第二个batch的时候会报错Trying to backward through the graph a second time,第一个batch的数据并不会报错。不知道作者是否遇到过类似的情况?
The text was updated successfully, but these errors were encountered: