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
GGML_ASSERT: D:\a\ctransformers\ctransformers\models\ggml/llama.cpp:453: data
I get this error sometimes when loading a model. At first, I thought it was a corrupted model, and I redownloaded it which fixed the issue. However, later, the issue reappeard, and the metadata of the file showed that it hasn't changed since I redownloaded it. After a lot of digging into the code of ctransformers, I concluded that it is probably a problem with not having enough memory, so I increased the swapfile and it resolved the issue completely.
Please do something to make this error more clear. Maybe throw a Python "MemoryError". That at least sends me in the right direction to finding a solution.
PS. I do understand that this error might be triggered in multiple ways, not just by a MemoryError. So, I suggest adding a line to the error that says something like "this could be a memory-error. Please make sure you have enough memory available."
The text was updated successfully, but these errors were encountered:
@jalajc, does increasing the size of the swapfile solve the issue for you too?
Yes.
Apparently I was on wsl2 Ubuntu jammy and the default settings didn't allow much of Ram.
But your comment thankfully hinted me to check memory inside, otherwise I was assuming I had enough memory.
GGML_ASSERT: D:\a\ctransformers\ctransformers\models\ggml/llama.cpp:453: data
I get this error sometimes when loading a model. At first, I thought it was a corrupted model, and I redownloaded it which fixed the issue. However, later, the issue reappeard, and the metadata of the file showed that it hasn't changed since I redownloaded it. After a lot of digging into the code of ctransformers, I concluded that it is probably a problem with not having enough memory, so I increased the swapfile and it resolved the issue completely.
Please do something to make this error more clear. Maybe throw a Python "MemoryError". That at least sends me in the right direction to finding a solution.
PS. I do understand that this error might be triggered in multiple ways, not just by a MemoryError. So, I suggest adding a line to the error that says something like "this could be a memory-error. Please make sure you have enough memory available."
The text was updated successfully, but these errors were encountered: