Jian Shi

Results 19 issues of Jian Shi

## Bug report - AirSim Version/#commit: master/1.7.0 - UE/Unity version: 4.25 - autopilot version: - OS Version: Windows 10 ### What's the issue you encountered? I was following the instructions...

bug-report

Adding mosaic augmentation for a better YOLO-series support. The codebase is expected to be finished by 20th May. Details of the augmentation may refer to [here](https://programming.vip/docs/yolov5-data-augmentation.html). Test code: ```python import...

I was thinking of refactoring the `augmentation` module with the following base for both `Augmentation`s and `Sequential`s. So that, all these abstract classes will be implemented by `IntensityAugmentation` and `GeometricAugmentation`...

help wanted

## ๐Ÿš€ Feature Optimizing of ```same_on_batch``` params for augmentations. ## Motivation All operations and random param generations are designed towards performing element-wise augmentations. With `same_on_batch`, it only controls the parameter...

enhancement :rocket:
help wanted
module: augmentations

#### Changes Fixes # (issue) #### Type of change - [ ] ๐Ÿ“š Documentation Update - [ ] ๐Ÿงช Tests Cases - [ ] ๐Ÿž Bug fix (non-breaking change which...

docs :books:

#### Changes Fixes # (issue) #### Type of change - [ ] ๐Ÿ“š Documentation Update - [ ] ๐Ÿงช Tests Cases - [ ] ๐Ÿž Bug fix (non-breaking change which...

Added the auto nn builder with a module registry. We can also add some custom field as a record like: ```python @nn.register_module(differentiable=True, jitable=False) class RandomFlip(...) ``` /cc @lferraz

enhancement :rocket:
module: nn

Remove the backend error to allow using other backends. e.g. error will occur if using plaidml.

I took an example from Plotly for 3D ploting but getting an error with: ``` Traceback (most recent call last): File "flexx_plotly.py", line 22, in flx.launch(PlotlyGeoDemo) File "/home/ovs-dl/anaconda3/envs/dl/lib/python3.6/site-packages/flexx/app/_funcs.py", line 231,...

type: question

Hi, Thanks for the repo. I can build the DCN with torch==1.5.1, but I got an error when I run ```python test.py``` ``` "test.py", line 5, in from model import...