David de la Iglesia Castro
David de la Iglesia Castro
It would be a nice feature to allow to serve MLEM models through https://developer.nvidia.com/nvidia-triton-inference-server . Afaik, it supports all the frameworks currently supported by MLEM.
It would be a nice feature to allow to serve MLEM models through https://pytorch.org/serve/ . This is only suitable for PyTorch models so I am not sure if it's out...
The API for `apply-remote` feels a little confusing to me, requiring to pass too much info and args: ```bash mlem apply-remote http test_x.csv -c host="0.0.0.0" -c port=8080 --json ``` Given...
Sorry if I'm missing some context regarding the scope of `mlem`; I have reading a bit of the existing documentation in Notion and skimmed through the code in this repository....
The current plots analyze the images and bounding boxes separately. There is some important insights missing that need to combine both sources of information. For example: - Total Area occupied...
# Motivation Other popular *ML Loggers* have methods for logging and tracking *hyperparameters*: - [Wandb](https://docs.wandb.ai/guides/track/config) - [MLFlow](https://mlflow.org/docs/latest/python_api/mlflow.html#mlflow.log_param) - [CometML](https://www.comet.ml/docs/python-sdk/Experiment/#experimentlog_parameter) - [Neptune](https://docs.neptune.ai/you-should-know/what-can-you-log-and-display#parameters-and-model-configuration) The existence of these methods are useful when building...
Should focus on answering and providing examples, as clear and soon as possible, for: - How to make it work with `DVC`. - How to make it work with `{X}`...
😓 Catalyst uses `valid` instead of `eval` so I actually forgot to include it in the standardization _Originally posted by @daavoo in https://github.com/iterative/dvc.org/pull/3726#discussion_r914724169_
In order to properly [resume training with dvc checkpoints](https://dvc.org/doc/user-guide/experiment-management/checkpoints#starting-from-an-existing-checkpoint), the user needs to load the existing `model_file` at the beginning of training. Given that DVCLive integrations already take care of...
The [TensorBoard callback](https://www.tensorflow.org/api_docs/python/tf/keras/callbacks/TensorBoard) collects and stores learning rate: https://github.com/keras-team/keras/blob/d8fcb9d4d4dad45080ecfdd575483653028f8eda/keras/callbacks.py#L2527-L2531