-
Hello.
I have an Intel ARC A380 and I'm using Ollama with IPEX-LLM using this script with Ubuntu:
```
#!/bin/bash
# Activate conda environment
source /home/nikos/miniforge3/etc/profile.d/cond…
-
What I'd love is, for tensors that are called with different shapes, is a sorted list by decreasing frequency. This would be very useful. However, `--verbose` didn't change the output?
-
Investigate whether using `Tensor` in backend operations can improve speed.
Relevant materials:
https://github.com/dotnet/core/blob/main/release-notes/9.0/preview/preview4/libraries.md#new-tensort…
-
When training the gimVI model, I am running into a ValueError in the first epoch for some datasets and not others. In all cases, the inputs are AnnData objects with raw counts for both the RNAseq …
-
I would expect these asserts to pass. The last assert fails with QCOM=1. They pass with CLANG=1 on PC. It's important to be able to create Tensors backed by numpy array so we can run a model quickly w…
-
### System Info
Latest TRL from source, can't run TRL env rn as cluster is shut down but I'm installing everything from source.
If required will restart cluster and run.
### Information
- [ ] Th…
-
Having been fiddling with the front-end loss function code and having been looking at mini-Vehicle for the paper, it's become apparent to me that we really should switch to tensors in the front-end. I…
-
We should accept shape tensors when setting up optimization profiles in the compiler. Additionally, the compiled function should provide some kind of mechanism to query shape bounds of its outputs.
-
I'm developing a peft algorithm, basically it does the following:
Say the training process has 30 steps in total,
1. For global step 0\~9: train `lmhead` + `layer_0`
2. For global step 10\~19: train …
-
running with --directml acceleration first got this error:
F:\ComfyUI\custom_nodes\ComfyUI-GGUF-main\nodes.py:79: UserWarning: The given NumPy array is not writable, and PyTorch does not support no…