Tarun Narayanan
Tarun Narayanan
Hey any update on releasing the training code for PP Matting?
Hi @wuyefeilin I did try doing that - I changed the outer `pretrained` key under the model key to the downloaded MODNet HR18 `.pdparams` file. But I got ```ValueError: paddle.load...
I'm looking for this as well @smallcube . any chance you found anything?
Hey, so all images and text files should be in one single folder?
Does this work `data/images/p1.jpg` and `data/text/p1.txt`
Hi I prepared my dataset in that structure and I ran the below command ```python train.py --model_name RN50 --folder /data/depop/data_org/clip/data/ --batch_size 512 --gpus 1``` I'm getting an `AssertionError` from the...
Okay so in `models/wrapper.py` is the `warmup_step` hardcoded to 2000? My dataset currently is much smaller for the num_training_steps to be bigger than 2000.
Same issue, I'm currently training SimCLR on the default settings given by you, and my loss has constantly been approx 30.00 (± 2) from epoch 1 to 90. What could...
@HobbitLong Yes, I've tried adjusting it between 0.1 to 1, the loss still is the same from epoch 1 to 100. Infact I'm just directly running the same code on...