mi-prometheus
mi-prometheus copied to clipboard
IBM
Enabling reproducible Machine Learning research
**Describe the bug** models/thalnet/thalnet_module.py is the source of the error, on line 131. Trying to log an error to logger causes this. **To Reproduce** Steps to reproduce the behavior: Happened...
There are still a few things to do to enhance further the documentation: - Explain the contents of the configuration files: the section (mandatory vs optional), parameters (mandatory vs optional)...
Take into account: - new flags for all workers/grid workers and - introduce helpers
There is inconsistent naming across params. For example, directories are sometimes called 'dir' and sometimes called 'folder'. We should perhaps decide on a unified set of standard names for source...
**Describe the bug** Why two test runs on MNIST test set with different random seeds return different values **To Reproduce** 1. Train a model: mip-online-trainer --c configs/vision/simplecnn_mnist.yaml --li 2. Run...
In current trainers/tester flow there are several loops that need to be broken if a given condition is true. In particular, some conditions need to be checked few times (e.g....
This is a worker which: - runs grid-trainer (and creates a new folder with experiments) - runs grid-tester on the resulting folder with experiments - runs grid-analyzer on the final...
Consider actually NOT stopping when "converged", this will require dividing loss_stop into two parameters: - loss_threshold (DEFAULT: 1e-5) - indicating the threshold - loss_stop (DEFAULT: True or False?) - indicating...
*validation with batch of size 1* - works perfectly *validation with batch of size 10* - from time to time returns values that differ First I was thinking that the...
General idea is to reproduce "tiny helpful apps", like e.g. index_splitter OR... grid workers(!)