-
I have CT dataset with high class imbalance for binary classification, where only 1% of my 3D CT scan is to be segmented. How do i choose learning rate, loss and epochs?
-
# Interview questions of ML algorithm
## Content
* Data validation:
> Data quality check, missing value, imbalance data & etc
* Modeling
> Common machine learning models
* Evaluation
> All …
-
Thanks so much for your implementation. But I have several questions:
1) In the below picture, it seems that the class with less numbers is sampled repeatedly, while the class with more numbers is s…
-
Problem Description:
Fraudulent transactions are a major threat in banking and insurance, causing substantial financial losses and damaging customer trust. Swift, accurate fraud detection is essentia…
-
### Problem: Rebalance with dask-cuda does not rebalance effectively
### Problem Context
We can not seem to rebalance with `dask-cuda` effectively which leads to imbalanced GPU usage.
This…
-
Add stress tests that simulate common/challenging allocation patterns. For example:
- Multi-threaded producer/consumer. The producer allocates jobs, and the consumers process them (perhaps doing som…
-
### Proposal Details
The current [strategy](https://cs.opensource.google/go/go/+/refs/tags/go1.22.5:src/database/sql/sql.go;l=1324-1330;drc=0d018b49e33b1383dc0ae5cc968e800dffeeaf7d) of lookup of a fr…
-
### Issue Description:
The current implementation of the `computeKernelBasisVectors` function assumes that every class or subset of samples in the dataset will have an equal number of samples. This a…
-
Using `num-part` with the `beta,0.2,0.2` distribution makes vnbest hang sometimes, esp with runs with more than 1000 weights.
-
It should be checked if the dropping chance of potion is more frequent.
If the percentage of dropping the potions is imbalanced, it should be reduced.
Depending on implementing https://github.com/…