-
I tried to run 'test_net.py' file, and I got the error:
return self._aggregate(person_boxes, query, person_key, object_key, hand_key, mem_key)
File "/media/data4/home/HIT/hit/modeling/roi_heads/ac…
-
Hi, I tried a test about compiling unet(torch.float16), which is the part of StableDiffusionXLPipeline in Inferentia2.8xlarge and it failed.
When the latent size of unet is (64, 64), it did not fai…
-
In many applications we need bmm for gram matrix calculation, like in neural style. However it seems gram matrix with 01 mode will always give NaN.
See the issue here
https://github.com/pytorch/p…
-
## Issue description
For fused_attention-related models like `ElectraForQuestionAnswering`, the fused_attention pattern could not be matched for dynamic shape due to `sym_size`. The failure of SDPA…
-
When I use `calculate_flops` to calculate flops of a local model (e.g. `openai/clip-vit-large-patch14-336` downloaded locally), the result is smaller than the FLOPs calculated manually (use the flops …
-
I'm getting this error message when I try to launch balatro mobile maker
-
Standalone MM 49.13
Windows 10
Java 11
This was a fresh install of MM 49.13. On the 'Advanced Rules' tab of the 'Game Options' window, the top checkbox is 'TacOps Minefield / BMM Minefield Suppor…
-
Data on the `dst` register in the compute kernel looks good, but is corrupted when printed from the output CB after packing -- hence, narrowed down to `matmul_pack_tile` function, which calls `llk_mat…
-
In
https://github.com/sjvasquez/instacart-basket-prediction/blob/master/models/blend/prepare_blend_data.py#L43
you look to load predictions.npy - but this file is never generated because "predi…
-
### 🚀 The feature, motivation and pitch
There is a forward implementation for `torch._native_multi_head_attention` but no coressponded backward implementation. So use torch to realize training MHA,…