-
**Description**
I've used triton to serve my model using `onnx_backend` run with `TensorRT FP16` but inference results are incorrect.
I've run with `TensorRT FP32`, then the result is correct.
…
-
This is the job script I was using for my parameterized experiments:
```shell
#!/bin/bash
#SBATCH --partition=cluster
#SBATCH --mem=80GB
#SBATCH --cpus-per-task=32
#SBATCH --time=06:00:00
#SB…
-
Hi gdlg:
I can not train the model wirh my own panoramic dataset. Moreover, i want to know how the training data is labeled? I need your help! Thanks!
-
I'm getting a segmentation fault running ofxDarknet on Ubuntu 18.0.4.
**...
13 conv 512 3 x 3 / 1 1 x 1 x1024 -> 1 x 1 x 512
14 conv 425 1 x 1 / 1 1 x 1 x 512…
-
Issue during saving `unsloth/mistral-7b-instruct-v0.3-bnb-4bit` after training/saving, both in Kaggle and [gguf-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo)
I have tried converting…
-
Hi, Could you please help me to make sure if the prototxt file is correct or not?
Thanks
[yolov4-tiny.txt](https://github.com/ChenYingpeng/darknet2caffe/files/4849163/yolov4-tiny.txt)
ai815 updated
3 years ago
-
It should be possible to maintain compatibility with most common models used in CV, so what do you think of having examples using segmentation with models directly from Metalhead. Also, it would be go…
-
I trained a custom EfficientNetB2 classification model with Tensorflow 2.9.0:
```
base_model = EfficientNetB2(include_top=False, pooling ='avg', weights='imagenet', input_shape=(IMG_SIZE, IMG_SIZE,…
-
Right now, if the user fits an estimator using a `pandas.DataFrame`, but passes a `numpy.ndarray` during `predict`, they get a warning due to missing feature names.
The situation is only to get mor…
-
### Proposal
I would like a simple `get_sae_config` that takes the `release` and `id` (under `saes`), and returns the config for the SAE.
This should not require downloading the whole SAE weights…