-
Conv kernels will handle non-tile multiple input shard height by leaving garbage rows in the tilized CB and output CB. Output shard can be allocated with padded/tile aligned size if conv output is til…
-
### Expected Behavior
-
### Actual Behavior
SD15 and Flux work fine, the problem is only with SDXL
Co…
-
It's me again : ) Thanks for your beautiful code , it's so much pretty and clearer than the offical version. The lost functions of this model seems not require ground truth. Is it a unsupervised deep-…
-
I try to run the TTS (English and Multi Language Text-to-Speech) in my PC.
https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/pipeline…
-
**Describe the bug**
I'm using 24 A100 (40G) video cards for llama-2-70B training, and previously had a lot of OOM issues with deepspeed ZeRO-3, so I'm currently using multi-card parameter parallel…
-
# Abstract
Aqua has the DOcplex translator for optimization problems. It automatically generates an Ising Hamiltonian for an optimization problem and you can solve it with quantum computers by applyi…
-
**Description**
I'm trying to serve an embedding model [FastText] in triton-server using python as its backend. The external dependencies are just fasttext module which is inturn dependent on numpy. …
-
**Gammapy version**
1.1
**Bug description**
Attempting to resample the reconstructed energy axis of a dataset yields unexpected results.
**Expected behavior**
For a dataset, make two copies. …
-
## Environment:
- TensorFlow Version: 2.15.0.post1
- Keras Version: 3.0.4
## Issue Description:
An error occurs when converting a Keras model to TensorFlow Lite format. This issue arises using …
-
The `anova` method for `merMod` objects always refits a LMM with `REML = TRUE` via `refitML.merMod`. This may lead to undesired consequences, as `refitML.merMod` does (a) neither inherit the optimizat…