Guodong Xu
I followed "Image Classification with Pre-trained Imagenet CNN" instruction and was stuck while trying to implement julia > net = Net("imagenet", backend, layers) it follows 26- 4 09:53:00:INFO:root:Constructing net imagenet...
Platform (like ubuntu 16.04/win10): **ubuntu16.04** Python version: **3.7** Source framework with version (like Tensorflow 1.4.1 with GPU): **mxnet** Destination framework with version (like CNTK 2.3 with GPU): **tensorflow 1.8 with...
rt, thanks~
Hi I have found that there is an intent to call julia from r in todo.md. There is a r package https://github.com/armgong/rjulia enabling us to call julia from r。 Is...
Hi, thanks for your work. I'm wondering is there a rough timeline for the code release?
Hi, I noticed in your paper the training epoch is 20 and warmup is 5 epochs, but when I tried to reproduce the results with train.sh, it runs for 100...
Hi, Thanks for your great works! I'm currently trying to reproduce EVA-02 Pre-training stage (Vit-B, 4 nodes x 8 40GB A100) and I find it takes 13~ hours to finish...
Great work!. I have requested the dataset but didn't get a reply, could you take a look? my email is [email protected] Thanks!
Hi, I'm currently trying to reproduce the results from a paper, but I've noticed some discrepancies with the paper's results. For COCO: ```bash CUDA_VISIBLE_DEVICES=0,1,2,3 vit=l task=ovd dataset=coco bash scripts/train.sh [10/29...
Hi Dear Author, Great work! I'd like to inquire where I can find the address for Audio-Language Alignment data. I noticed in `scripts/audio_language/train.sh` that there is a mention of 4,800,000...