-
The script should:
* allow loading a pre-trained agent, an image dataset as well as a set of ground truth bounding boxes
* let the agent act on the images from the dataset, logging its predicted bou…
bakoe updated
5 years ago
-
I found that `chainer==6.0.0b2`breaks my code.
I'm using a helper class that contains a numpy/cupy array and do something with it. The device of this array can be switched by `to_cpu/to_gpu` method…
-
There is a error when I tried to train my model:
I checked the chainer's code, I found no SSD300Plus but a similar class named "SSD300" in chainercv/links/__init__py. Is it the right one?
-
Can I output the log using the function of report in `chainercv/examples/semantic_segmentation/eval_semantic_segmentation.py` ? Is it possible to use the reporter function without relying on the train…
Keiku updated
5 years ago
-
Hello,
I trained VOC using the default settings, and it completed in 16 hours on a single GPU.
Now i tried training my own custom DB which has only 1500 images and a single class however the time …
ghost updated
5 years ago
-
![image](https://user-images.githubusercontent.com/17477947/50586727-da356780-0ea0-11e9-8219-05b45856cd82.png)
This is what I am getting when I run python calc_weight.py, an EOF error, what could b…
-
Can i use the ChainerCV to detect only the objects that belong to single class, like i want to detect only persons face and draw a bounding box around that face using frcnn or ssd? is it possible?.
-
A training loop got stuck in a certain condition with multi-processing updater and opencv. The issue does not appear when I use Pillow or serial updater.
* Conditions
- Chainer version: 2.0
…
-
The problem is that `chainercv` uses [`utils.read_image`](https://github.com/chainer/chainercv/blob/master/chainercv/utils/image/read_image.py) function to load an image, where users can set the optio…
-
I can't find a function define in ChianerCV.But your code use this function in eval_model.py.Please tell me how to deeal with it.