-
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 …
-
Hi, I'm seeing performance accuracy drops if I try to compile (via edgetpu_compiler) a mobilenet V2 model for image classification.
When I use the pre-compiled MobileNetV2 model provided at https:…
-
**System information**
- Have I written custom code (as opposed to using a stock example script provided in TensorFlow): Yes
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 16.04…
-
![slimmable_networks](https://user-images.githubusercontent.com/22609465/50390872-1b3fb600-0702-11e9-8034-d0f41825d775.png)
-
I try to test the TfLiteMnasNet-Armnn with [MnasNet_1.3_224](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/guide/hosted_models.md) but get a weird output. Could it be poss…
-
**1) Order with cnn, relu, batch norm**
https://github.com/azamatkhid/mnasnet-pytorch/blob/f855edc8d114636f88ef586fa257a7f5d1f52b9b/model.py#L30-L32
CNN -> Relu -> Batch Normalization
but ref…
-
We've realized that the hashing of properties in the config hides things from
users and makes using DFFML confusing.
We should make the directory property required so that users always know where
the…
-
I want to use NAS method to search a resource-aware model, but for each step, the mutator generates a new model structure so that it is not easy to calculate the generated model size, especially flops…
-
hi, I built the Tengin in Mac OS and Ubuntu 18.04 separately, and I test it in rk3399 use `adb` successfully , but I got the error like followings in both two platform when I try to convert the caffem…
-
Hi @mingxingtan, I have a question regarding the comparison of EfficientNet with other architectures in Table 2 of the [EfficientNet paper][1].
It does not seem to be mentioned in the paper but is …