junos
/
rapids
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Save cleaned features with different data cleaning threasholds under different folders

pull/95/head
Meng Li 2020-05-15 17:59:14 -04:00
parent 78cd8159be
commit 6e99e69f8c
1 changed files with 12 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
rules/models.snakefile
View File

@ -61,13 +61,13 @@ rule clean_features_for_individual_model:
input: input:
rules.merge_features_for_individual_model.output rules.merge_features_for_individual_model.output
params: params:
cols_nan_threshold = config["PARAMS_FOR_ANALYSIS"]["COLS_NAN_THRESHOLD"], cols_nan_threshold = "{cols_nan_threshold}",
cols_var_threshold = config["PARAMS_FOR_ANALYSIS"]["COLS_VAR_THRESHOLD"], cols_var_threshold = "{cols_var_threshold}",
rows_nan_threshold = config["PARAMS_FOR_ANALYSIS"]["ROWS_NAN_THRESHOLD"], days_before_threshold = "{days_before_threshold}",
days_before_threshold = config["PARAMS_FOR_ANALYSIS"]["PARTICIPANT_DAYS_BEFORE_THRESHOLD"], days_after_threshold = "{days_after_threshold}",
days_after_threshold = config["PARAMS_FOR_ANALYSIS"]["PARTICIPANT_DAYS_AFTER_THRESHOLD"] rows_nan_threshold = "{rows_nan_threshold}",
output: output:
"data/processed/{pid}/data_for_individual_model/{source}_{day_segment}_clean.csv" "data/processed/{pid}/data_for_individual_model/{rows_nan_threshold}|{cols_nan_threshold}_{days_before_threshold}|{days_after_threshold}_{cols_var_threshold}/{source}_{day_segment}_clean.csv"
script: script:
"../src/models/clean_features_for_model.R" "../src/models/clean_features_for_model.R"
@ -75,13 +75,13 @@ rule clean_features_for_population_model:
input: input:
rules.merge_features_for_population_model.output rules.merge_features_for_population_model.output
params: params:
cols_nan_threshold = config["PARAMS_FOR_ANALYSIS"]["COLS_NAN_THRESHOLD"], cols_nan_threshold = "{cols_nan_threshold}",
cols_var_threshold = config["PARAMS_FOR_ANALYSIS"]["COLS_VAR_THRESHOLD"], cols_var_threshold = "{cols_var_threshold}",
rows_nan_threshold = config["PARAMS_FOR_ANALYSIS"]["ROWS_NAN_THRESHOLD"], days_before_threshold = "{days_before_threshold}",
days_before_threshold = config["PARAMS_FOR_ANALYSIS"]["PARTICIPANT_DAYS_BEFORE_THRESHOLD"], days_after_threshold = "{days_after_threshold}",
days_after_threshold = config["PARAMS_FOR_ANALYSIS"]["PARTICIPANT_DAYS_AFTER_THRESHOLD"] rows_nan_threshold = "{rows_nan_threshold}",
output: output:
"data/processed/data_for_population_model/{source}_{day_segment}_clean.csv" "data/processed/data_for_population_model/{rows_nan_threshold}|{cols_nan_threshold}_{days_before_threshold}|{days_after_threshold}_{cols_var_threshold}/{source}_{day_segment}_clean.csv"
script: script:
"../src/models/clean_features_for_model.R" "../src/models/clean_features_for_model.R"