Andrei Moraru

@tom-pollak indeed best solution so far

@morestart, you probably already know the answer. However, in case anyone else was wondering. [LSTM](https://pytorch.org/docs/stable/generated/torch.nn.LSTM.html) in pytorch is a multi layer network, that is why you can select the number...

@thanhtvt Exactly! And this is precisely the example PyTorch provides in the docs: If you take a look at the [LSTM page](https://pytorch.org/docs/stable/generated/torch.nn.LSTM.html): ```python rnn = nn.LSTM(10, 20, 2) # (10...

> encoder_dim = encoder_out.size(3) Dimension out of range (expected to be in range of [-2, 1], but got 3) @yuhua666 , I am assuming you are not training in batches,...

That's an interesting point, but but I think this is covered. Notice how the center-encoded coordinates are also in percentages:  So clamping in between 0 and 1 should work...

I see your point. You might be onto something. Indeed, if you generate a prior box that is centered close enough to one corner of the image, you could easily...

The link to the torch checkpoint works from Chrome. Google drive is always iffy on other browsers for reasons, @NicolasHarding. Also, @sgrvinod, I know you made these tutorials years ago,...

@haoren55555, unfortunately the author is MIA, although he left three awesome complete tutorials. To your point, if I get your question correctly, yes, you do need to generate specific priors...

Hi, First of all great project! Really. I made a post on the Luxonis forum asking about the status of DepthAI for Unity and now I came here to see...

I have this issue as well on CUDA `11.7` and cudnn `8.5` with various tf versions (tried `2.10`, `2.8` and `2.6`, with `2.10` being the latest one that can run...