-
hi, i am an 4th year undergraduate, involved in an project where i am implementing customized superpoint (pytorch) based SLAM on KRIA KR260,
I am struggling to do quantizing and pruning using Vitis …
-
## ❓ General Questions
As I understand it, Half-Quadratic Quantization (HQQ), is a new technique for quantizing models to reduce the memory requirements of these models, making it easier to deploy …
yieme updated
7 months ago
-
We aim to implement a system that leverages distillation and quantization to create a "child" neural network by combining parameters from two "parent" neural networks. The child network should inherit…
-
Hi everyone,
I'm trying to quantize the YOLOv5n model from [here](https://github.com/ultralytics/yolov5). I'm using the Vitis-AI v3.0 docker with the following code:
```
import pytorch_nndct
i…
-
Getting this error after on quantizing after fine tuning with the instructions for colab.
**FileNotFoundError: Could not find tokenizer.model** in llama-2-7b-meditext or its parent; if it's in ano…
-
like in Jmask some sort of quantizing, a grid
I guess only the Random Note Processors could have one, as part of the Seed option
I think it would make the Random system complete
-
Hi team, I am having issue quantizing the network consisting of Conv and Linear layers using **int8** weights and activations in ONNX. I have tried setting it using op_type_dict, however it doesn't wo…
-
### Feature request
Hi! I’ve been researching LLM quantization recently ([this paper](https://arxiv.org/abs/2405.14852)), and noticed a potentially improtant issue that arises when using LLMs with 1-…
-
In our codebase we use the Router contract to check whether a timestamp or a block is bigger, smaller, equal (etc.) compared to the current value from context. It will be common that this value will b…
-
Seeing this for both qint8 and qint4 for flux. I saw that it was fixed in master. Would be nice if you could do a release with the fix :)
```
.venv/lib/python3.10/site-packages/optimum/quanto/tens…