-
Hi, I'm trying to implement your novel loss function in my dataset.
One thing about your ASL paper is confusing me, that is, which of the following pretrained model you used to implement ASL loss in…
-
GENet look promising in the paper. Better or similar results as EfficientNet/MobileNet with faster inference time.
Pytorch implementation: https://github.com/idstcv/GPU-Efficient-Networks
It compose…
-
Hi, I can't find the training script.
Thanks.
-
I think you made a mistake in "tresnet.py", line 265 and line 271.
you used 'tresnet_m_448' for tresnet_l_448 and tresnet_xl_448 versions, which leads to an error when loading pretrained models.
```…
-
Hello, I'm very interested in your novel work and try to run your validate.py file, but get totally wrong mAP:
Actually, I run
python validate.py \
--model_name=tresnet_l \
--model_path=./model…
-
RuntimeError: Given groups=1, weight of size [76, 48, 3, 3], expected input[1, 64, 112, 112] to have 48 channels, but got 64 channels instead
Actually it is giving error on small PNG files, on the …
-
Typically, ResNets are trained on 224 x 224 images.
Is there a pre-trained model on larger images like 720*720?
Any suggestions?
-
Hi,
I am currently working with the TResNet model, training with mixed precision. Regarding the IABN model, for Nvidia apex O2 level, it is needed to recast the module to float32. Is this also nece…
-
Is it correct ?
```
julia> tresnet(tensor(rand(Float32,224,224,3,1), dev=0))
1000×1 Tensor{Float32,2}:
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
…
-
Since ReXNet is based on MNASnet architectures(mobile et v1 and v2) I guess that they suffer from the same low throughput low GPU performance issue.
Can you provide any numbers?
I am specifically …