MPC-58 integrate training with mlflow #38
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dnerini
commented
Nov 19, 2024
| @@ -75,6 +75,13 @@ def main(): | |||
| help="If set to 1 (default), the input parquet files (homogeneized tables) for the ml routines will be recomputed from the current database rows"+ | |||
| "This takes a bit of time but is needed if you updated the database and want to use the new data in the training", | |||
| metavar="MODELS") | |||
|
|
|||
| parser.add_option("-l", "--logmlflow", | |||
There was a problem hiding this comment.
we need a flag to decide whether to log the model artifact, false by default?
There was a problem hiding this comment.
parser.add_option("-l", "--logmlflow",
type="choice",
choices=["none", "metrics", "all"],
dest="logmlflow",
default="none",
help="Specify the logging mode for MLFlow. Choices are:" +
" 'none' (default, no logging), 'metrics' (log metrics only)," +
" or 'all' (log metrics and model)." +
" To log to a remote ML server, the environment variable MLFLOW_TRACKING_URI needs to be set.")
How about a choice like this?
dnerini
commented
Nov 19, 2024
| import os | ||
| from scipy.interpolate import UnivariateSpline | ||
| from pathlib import Path | ||
| import mlflow |
There was a problem hiding this comment.
should we make mlflow an optional dependency?
for example
try:
import mlflow
MLFLOW_INSTALLED = True
except ImportError:
MLFLOW_INSTALLED = False
There was a problem hiding this comment.
good idea, we will change that
|
I just added the possibility to log also test errrors to mlflow., using cross-validation. Sorry for the stupid commit name. It works by using the argument -C <number_of_crossval_iterations> in rf_train.py. Default is 0 : no test error, no cross-val |
…b.com/MeteoSwiss/rainforest into MPC-58-Integrate-training-with-MLflow
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.