Leon Truong
Here's my experimentation with and without tanh in the encoder. Note, I've ensured I've set my model to eval + no_grad before predicting and no_grad during validation. which is different...
did you install tensorflow with GPU? . it should use GPU by default. `pip install tensorflow-gpu`
@AshleyRoth you can limit the gpu memory using the code below. Add it to the script which you are invoking. Keeping in mind you must allocate enough for the model...
@harshil014 I was able to use the latest cuda version that tensorflow supports.
@Shreeyak yes.
python 3.6. I'm using it in virtualenv
For anyone else having trouble with this, this was my workaround. 1. Extend the predict handler like so. ``` import kserve.handlers as handlers class CustomHandler(handlers.PredictHandler): def set_default_headers(self, *args, **kwargs): #...
I got warp terminal working on windows 11 using the suggested install. This uses window 11's GUI apps. [reading](https://github.com/microsoft/wslg) 1. Download warp `.deb` install from warp website https://app.warp.dev/download?package=deb 2. Install...
@Welding-Torch thank you. I didn't realise the copy and paste was ctrl+shift+V. I went into settings to change the default back to ctrl+C and ctrl + V for copy and...
@vaishnavm217 to fix this do we just do something like this? to normalise the bounding boxes. ```python w,h = image.size boxes /= torch.Tensor([w,h,w,h]).expand_as(boxes) ```