-
### Proposal
## Problem
I was discussing this problem with @ArthurSens on CNCF slack, and seems like CT zero-samples scraped by Prometheus in Agent mode do not make into remote write requests. Upo…
-
I need help with codes which would help classify the samples I have analysed using ATR-FTIR spectroscopy. I cannot set the target as my Excel file has 1667 columns and 80 rows. All columns have numeri…
-
-
Here are some good samples worth considering, and I think they are licensed under a free license.
* https://github.com/sonic-pi-net/sonic-pi/tree/dev/etc/samples
Please add other options to this t…
-
### Context
The current way that we generate computed files for multiplexed samples was developed with the expectation of a multithreaded environment with all computed files being generated on the …
-
# The Issue
SF3 uses Ogg Vorbis compression, which is lossy. This poses a problem because loading an SF3 into Polyphone and saving it again decompresses and then recompresses the samples, leading to …
-
I tried running the model using the command `python main.py --input_dir debug/ --input_name samples.json` on 3 samples constructed from the examples in the paper's appendix.
It gives a score of 5 …
-
For the first run, the Evaluation in step 200,400,800:
{'eval_loss': 0.6967583894729614, 'eval_accuracy': 0.5025337837837838, 'eval_f1': 0.3344575604272063, 'eval_matthews_correlation': 0.0, 'eval…
-
## Summary
Poisson distribution where, at high lambda values, the sample variance deviates noticeably from the mean.
During a Kolmogorov-Smirnov (KS) test, the [Poisson distribution fails](https…
-
### Description of feature
Would be nice to have a single object instead of inidividual h5ad files.
Optional batch correction for integration.