Redo experiments with fixed dataset #71
Merged
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
…riments to be run tonight
… on linux machine
danyoungday
commented
Mar 13, 2024
use_cases/eluc/data/conversion.py
Outdated
| if old_abbrev in MANUAL_MAP.keys() and MANUAL_MAP[old_abbrev] in codes_df["Numeric code"].unique(): | ||
| countries_df.iloc[i]["abbrevs"] = codes_df[codes_df["Numeric code"] == MANUAL_MAP[old_abbrev]]["Alpha-2 code"].iloc[0] | ||
| countries_df.loc[i, "abbrevs"] = codes_df[codes_df["Numeric code"] == MANUAL_MAP[old_abbrev]]["Alpha-2 code"].iloc[0] |
There was a problem hiding this comment.
Was giving me a pandas warning because the old way will be deprecated soon.
danyoungday
commented
Mar 13, 2024
| @@ -169,6 +169,7 @@ def __init__(self, start_year=1851, test_year=2012, end_year=2022, countries=Non | |||
|
|
|||
| self.train_df = df.loc[start_year:test_year-1] | |||
| self.test_df = df.loc[test_year:end_year-1] | |||
| assert self.train_df['time'].max() == self.test_df["time"].min() - 1 | |||
There was a problem hiding this comment.
Some assertions to make sure the same mistake doesn't happen again
danyoungday
commented
Mar 13, 2024
There was a problem hiding this comment.
Reran experiments with fixed dataset
danyoungday
commented
Mar 13, 2024
There was a problem hiding this comment.
Moved predictor significance portion of notebook to its own python script
…updating with new trained TorchPrescriptors
ofrancon
approved these changes
May 14, 2024
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# Note: The original paper trains from 1982 onwards but this is too slow and large for the\n", | ||
| "# purpose of this example.\n", | ||
| "forest.fit(dataset.train_df.loc[2002:][constants.NN_FEATS], dataset.train_df.loc[2002:][\"ELUC\"])\n", | ||
| "forest.save(\"predictors/sklearn/trained_models/experiment_rf\")" | ||
| "forest_year = 1982\n", |
There was a problem hiding this comment.
In the comment above you say training from 1982 "is too slow and large for the purpose of this example". But now you do it. Should you remove the comment?
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "forest.load(\"predictors/sklearn/trained_models/experiment_rf\")\n", | ||
| "# TODO: I don't think we can possibly load a model this big\n", | ||
| "# forest.load(\"predictors/sklearn/trained_models/no_overlap_rf\")\n", |
There was a problem hiding this comment.
You saved the model a few cells above. So it's not too big to be loaded. Should you remove the TODO?
Fully trained torch prescriptors and updated experiments
… so it should be fine
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.
Previously the dataset overlapped the year 2012 in the train and test set. This redoes all the results in the notebooks with the updated dataset. Additionally, predictor significance is moved to a script rather than notebook.