njuhang

memory info doesn't match

3

The previous display total used memory plus the middle tensor memory does not equal the following total used memory

What does 'splits' mean

1

inception model takes in 3 channels input, how to compute IS on single channel image?

inception model takes in 3 channels input, how to compute IS on single channel image?

operands could not be broadcast together with shapes(255,32,30)(255,32,2)

3

在进行evaluate的时候，执行boxes[...,2:] = boxes[...,2:] - boxes[...,:2]时遇到这个问题 When evaluating, I encountered this problem when executing ‘boxes[...,2:] = boxes[...,2:]-boxes[...,:2]’

Can MMD be used to measure the performance of a generator？

1

I trained a generator, hoping to measure performance by MMD loss between the original data set and the generated data set. Is it feasible? How to calculate the MMD of...

error when using llm notebook with aimet pro

1

Hello, I followed instructions in tutorial_for_llama3p2-3b_iot(v1.0.0.2) and tutorial_for_Qwen2.5-1.5B(v0.1.0.0), but I encounterd the same problem when I export the quantized model: Traceback (most recent call last): File "/opt/notebook_uv/llama_venv/lib/python3.10/site-packages/torch/onnx/_internal/exporter/_capture_strategies.py", line 118, in...

bug on LSTM

5

I followed instructions in lpai4.x sdk, CRNN example. The model is defined as: class CRNN(nn.Module): def __init__(self, num_classes=12, dropout_prob=0.1, is_bidirectional=False): super().__init__() self.n_maps = 48 self.fc_maps = 84 # input NCHW...