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
After few days trying, eventually i can run clip/run_zero_shot_eval.py, and I am happy with the result by 4090.
However, when it comes to clip/run_pretrain_dist.py to do my custom sft, the code doesn't work and pop up error message like this:
File "/home/heng/anaconda3/envs/paddlemix/lib/python3.10/site-packages/paddle/tensor/math.py", line 974, in multiply_
_dygraph_tracer().has_grad is False
AssertionError: The current inplace version of multiply needs to be used in the context of paddle.no_grad() since inplace multiply_grad is not yet supported.
is there anyone met this just like what happens to me?
thanks for your time!
The text was updated successfully, but these errors were encountered:
hey guys, I am quite new to paddlemix :)
After few days trying, eventually i can run clip/run_zero_shot_eval.py, and I am happy with the result by 4090.
However, when it comes to clip/run_pretrain_dist.py to do my custom sft, the code doesn't work and pop up error message like this:
File "/home/heng/anaconda3/envs/paddlemix/lib/python3.10/site-packages/paddle/tensor/math.py", line 974, in multiply_
_dygraph_tracer().has_grad is False
AssertionError: The current inplace version of multiply needs to be used in the context of paddle.no_grad() since inplace multiply_grad is not yet supported.
is there anyone met this just like what happens to me?
thanks for your time!
The text was updated successfully, but these errors were encountered: