-
### Describe the bug
> RuntimeError: The size of tensor a (154) must match the size of tensor b (2304) at non-singleton dimension 1
### Reproduction
```python
# StableDiffusion3Pipeline
pipe.enab…
-
there is a join operation going on here
```
# Include src and dst entity table features for `test_df`
test_df = pd.merge(
test_df,
src_entity_df,
how="left",
left_on=left_enti…
-
# Problem
Currently stateIDs and sailboatIDs are generated using 2 separate uint32 variable. however consider making one central UniqueID generator with a more complex and scalable architecture.
…
-
### Suggestion Description
Started using torchlearn to train models in pytorch using my gfx1100 card but get a warning that 1toch was not compiled with memory efficient flash attention.
I see ther…
-
- [ ] profile the usage of memory
- [ ] Array / Record are suspicious to use hashmap. In the regular usage, they mostly use a fixed length array. Need to confirm and figure out a new way of memory us…
-
Namespace(confidence_threshold=0.2, config_file='configs/LVISCOCOCOCOSTUFF_O365_OID_VGR_SA1B_REFCOCO_GQA_PhraseCut_Flickr30k/ape_deta/ape_deta_vitl_eva02_clip_vlf_lsj1024_cp_16x4_1080k.py', input=['…
-
# ❓ Questions and Help
memory_efficient_attention fw produce inconsistent results
not sure what was going on? incorrect built? some specific versions combinations?
for some combinations:
xfo…
-
ghost updated
2 weeks ago
-
I am interested in your latest work "Time-, Memory- and Parameter-Efficient Visual Adaptation",but I can't find the repo! Could you tell where is it? Thkan you!
-
### Is there an existing issue for the same feature request?
- [X] I have checked the existing issues.
### Is your feature request related to a problem?
```Markdown
Yes. Many oom issues.
``…