-
목표: 용희님이 주시는 데이터를 가지고 mobilenet v3 / shufflenet v2를 훈련해서 평가한다
필요한 자료: @hayleyshim
- 훈련 데이터
- 평가 데이터
조사해야 할코드 (혹시 사용하시는 코드가 있으시면 남겨주세요 )
@yesolyun
- mobilenet v3 훈련코드, 평가코드 (예솔님)
@H-YUR…
-
-
Since `faster_rcnn_mobilenet_v1_feature_extractor` is already provided, adding mobilenet_v2 and mobilenet_v3 should be easy. Mobilenet_v2 and v3 provides better accuracy than v1.
I've written feat…
-
We can use [group] to implement mobilenet and mobilenet-v2, however can we implement the mobilenet-v3 in existing darknet?
-
There are testing code for the both ssdlite mobilenet v2 and ssd mobilenet v3 but where are the training code for these?
-
```
weightsPath = "frozen_inference_graph.pb"
configPath = "ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt"
model = cv2.dnn_DetectionModel(weightsPath,configPath)
```
![image](https://user-im…
-
- Project needs to use [torchvision 0.11.0](https://pytorch.org/vision/0.11/models.html) where it provides follwowing pretrained models.
```python
import torchvision.models as models
mobilenet_v3…
-
Hi,
I generated a model from `MobileNet-V3` by mutating some layers. When run with TVM, the model produces lower accuracy than PyTorch. However, I expected the results to be exactly same.
To rep…
-
### Description of the bug:
Given a module that calculates the maximum of a tensor, the convert method failed to convert the model appropriately.
```
import ai_edge_torch
import torch
class Ma…
-
### bug描述 Describe the Bug
在一些提供构建模型的API中,设计了参数`scale`作为调整模型通道数的放缩因子。
因此,`scale`应该是一个非负数才使得放缩有意义。
但部分`scale`缺少参数验证,当输入负值时,程序仍然能够运行、输出。
这些API包括:
```
paddle.vision.models.MobileNetV2
paddle…