Francis Lata

UNet3D MLPerf

24

# Overview This is the training script for the UNet3D MLPerf model. # How to run training script ## Download dataset Run the following script from the root folder of...

Recommended text or phoneme tokenizer to use

3

Hi @cantabile-kwok, in the paper, there was not any recommended text or phoneme tokenizer to use. Do you have recommendations of what to use? Thank you.

Possible collaboration on CTXtxt2vec

3

Hi @cantabile-kwok, I’ve been chipping away on the unofficial implementation of the UniCATS paper [here](https://github.com/francislata/unicats). Since the second part is out and it sounds like you’re working on the txt2vec...

Update UNet3D inference to use MLPerf's evaluation set

5

# Inference set changes I noticed that MLPerf's [inference set](https://github.com/mlcommons/inference/blob/87ba8cb8a6a4f6525f26255fa513d902b17ab060/vision/medical_imaging/3d-unet-kits19/meta/inference_cases.json#L44) includes one extra file, `case_00400`. I did see that the official [kits19 dataset](https://github.com/neheller/kits19/tree/master/data) doesn't have this particular case. To compensate...

tinytqdm write support

1

**Overview** This PR adds support for `write` that is commonly used in quite a few [examples](https://github.com/tinygrad/tinygrad/blob/master/examples/mlperf/model_train.py#L230). Here's an example of it after running the following: ```python from tinytqdm.helpers import tqdm...