-
It's cumbersome that we don't provide an argument that people can use to manually specify a threshold in the predict() function. If the user knows what threshold they want to use, then we should make…
-
Optimize the max_cands in https://github.com/HEP-KBFI/ml-tau-en-reg/blob/main/enreg/config/model_training.yaml#L39:
- for the signal-only tasks jet_regression and dm_multiclass, 16 seems perfectly fi…
-
The validator scripts are not working on any datasets, seems to be an issue with the `fit` method and the subsequent LabelBinarizer call. This is the error when running the example code:
```
Typ…
-
This is a proposal and suggestion in case "MediaVision" rejects to implement services using GStreamer/NNStreamer pipelines.
If "MediaVision" accepts the proposal, "MediaVision" may refer to this docu…
-
Multi-class log-loss has an `epsilon` value which thresholds the input probability between `epsilon` and `1.0`. This causes multi-class log-loss to never be `Infinity`.
The binary log-loss does not…
-
Hello teacher, can you take a look at the structure format of your data set? Are the labels saved in excel? There are 4 annotations in total.Are all image formats TIF format?
y = np.array([2,2,1,66,54,23])
z = np.array([4,5,7…
-
Hi! I defined my custom_metric function following your documentation:
`def custom_metric(X_val, y_val, estimator, X_train, y_train, weight_val=None, weight_train=None):
start = time.time()
…
-
Thank you very much for your contribution!
1. I would like to ask why the IoU of one of the types has always been 0 when I use pointnet++ to do my own point cloud data semantic segmentation (binary…
-
In calculating gradients, the gradient of the softmax function is not calculated using the formula that is derived in the lecture notes. It seems like in the code, this step is skipped over, and the g…