-
#### Describe the bug
Cannot create an in-learning minimisation problem using ExponentiatedGradient and MLPClassifier.
The problem lies in __lagrangian.py_. The function __call_oracle_ insists on …
-
Hello, the accuracy of the results I obtained by running on the PISC data set is only about 60%. I set the training parameters by installing the parameters in the article. May I ask which correspondin…
-
Using the latest master, I'm noticing big improvement from 0.0.60. The output form the upscaling unet isn't nearly as "swirly", but I am noticing red or green bits of noise on the output images:
(…
-
Hi Danial,
I followed your instructions for setup on MacOS Monterey but ran into the following error while running the fast-bleu example code:
```python
from fast_bleu import BLEU, SelfBLEU
ref1…
-
Example (executed regardless of the value in the previous if-block at L174) : https://github.com/AliciaCurth/CATENets/blob/bb4da46a8bf139b60822755aaa200dcbb7e8a3ac/catenets/models/torch/slearner.py#L1…
-
from section 1.9, Suppl. Material
> "To decrease the relative importance of short sequences, we multiply the final loss of each training example by the square root of the number of residues after cro…
-
### Type of report
- [x] bug
- [ ] feature request
### Source Document
13.3 Removing nodes:
https://documentation.suse.com/ses/7/single-html/ses-admin/#salt-node-removing
### S…
-
Currently we only have the weights for B comb. bkg. as the unfolding weights are different between nominal and B USB region.
For `D*`, we need to have different weights for events IN and OUTSIDE th…
-
This was reported by Rick van Dam while using the shiny app.
and the following code:
```
library(ssdtools)
library(readr)
# read dataset
# the file argument of read_csv() assumes the file is…
-
In the bbox generation model, you use NLLLoss to comput the loss of label prediction, but before use NLLLoss you just use softmax function without torch.log, so the value of lloss is always a negative…