Nguyen Nguyen Anh
Nguyen Nguyen Anh
It is getting boring with portrait-centric, too simple dataset. Let move on with the non-portrait one, extremely perplexed dataset like furniture of all kind etc.. Steve
This is my original setup StyleGanV2-ADA. Even at 24,000 kimg (24 million images went through network). - Total images 47,000 (256x256) with 19 conditional classes - 4xV100 batch 64. -...
Is there an example for use named pipe for Audio (raw) and Video (raw) and create a hls output from Python ? Both inputs (audio and video) are all raw...
Google Cloud platform standard VM - 2 vCPU 2G RAM - Static internal IP - Static external IP - Ubuntu 18.04, clean from GCE nothing changed. Install, YES to all,...
Can someone make this code work with **tf.distribute.Strategy()** for multi GPU training? I simply gave up really as errors after errors. Thanks, Steve
I use the **yolo3_one_file_to_detect_them_all.py** for a 608x608 image. It was so slow that I have to time the entire your prediction process. It turned out that that **do_nms** takes 5-7...
Your codes works amazingly well. However, use Spectral Normalization will converge VERY fast and nice ! That is all. I changed 2 things: (1) Added Spectral Normalization to Conv2D layers...
1) Ubuntu 16.04 2) Docker tensorflow/tensorflow 1.13.1 and tensorflow/serving:latest-gpu 3) NVIDIA TensorRT 5.0.2 (https://docs.nvidia.com/deeplearning/sdk/tensorrt-install-guide/index.html) 4) Tensorflow object detection Faster RCNN Resnet101 successfully built w/ 2 classes only 5) Model are...
These 2 new networks are complicated by all means. I love the style of [eriklindernoren](https://github.com/eriklindernoren) as it is simple to understand. By any chance, anyone or **eriklindernoren** would post similar...
Is there any chance that this repo will support Tensorflow v2 ? Thanks, Steve