Lars Heimdal

pip install -e . fails

1

Thanks for your awesome work! Hi, I get the following error when building: ``` -- Generating done -- Build files have been written to: /home/larsh/development/pixel-perfect-sfm/build/temp.linux-x86_64-cpython-39 [ 56%] Built target pixsfm...

Keypoint score

1

All keypoints seems to have the same score/confidence, which means all keypoints are used even when it's impossible to know the location. e.g. due to occlusion. Is this the expected...

Model with stack does not work with int8 target type

15

Converting this dummy model with quantize_target_type="int8" and per_tensor=True throws an error in tflite ```python import torch.nn as nn import torch from tinynn.graph.quantization.quantizer import PostQuantizer from tinynn.converter import TFLiteConverter class StackModel(nn.Module):...

RMS Norm doesn't seem to be supported

5

Hi, converting a model that uses `nn.RMSNorm` does not work: ```python class RMSNormModel(nn.Module): def __init__(self): super().__init__() self.norm = nn.RMSNorm(3, 0.1) def forward(self, x): x = x.transpose(1, 3) # [N, H,...

Hi, I get these warnings when running an int8 quantized model on I.MX 8M Plus using the vx delegate: W [HandleLayoutInfer:332]Op 162: default layout inference pass. W [HandleLayoutInfer:332]Op 56: default...