Makefile.logreg1

include Makefile.feature.esb2

ALGO_NAME := logreg
C := 1.0
REGULARIZER := l2
CLASS_WEIGHT := balanced
SOLVER := lbfgs
MODEL_NAME := $(FEATURE_NAME)_$(ALGO_NAME)_$(REGULARIZER)_$(C)
METRIC_VAL := $(DIR_METRIC)/$(MODEL_NAME).val.txt

PREDICT_VAL := $(DIR_VAL)/$(MODEL_NAME).val.yht
PREDICT_TST := $(DIR_TST)/$(MODEL_NAME).tst.yht
SUBMISSION_TST := $(DIR_SUB)/$(MODEL_NAME)_sub.csv

all: validation submission
validation: $(METRIC_VAL)
submission: $(SUBMISSION_TST)
retrain: clean_$(ALGO_NAME) submission

submit: $(SUBMISSION_TST)
	kaggle competitions submit -c $(COMPETITION) -f $< -m $(MODEL_NAME)

$(PREDICT_TST) $(PREDICT_VAL): $(FEATURE_TRN) $(FEATURE_TST) $(CV_ID) | $(DIR_VAL) $(DIR_TST)
	python ./src/train_predict_logreg1.py --train-feature-file $< \
                                         --test-feature-file $(word 2, $^) \
                                         --predict-valid-file $(PREDICT_VAL) \
                                         --predict-test-file $(PREDICT_TST) \
                                         --C $(C) \
                                         --regularizer $(REGULARIZER) \
                                         --class_weight $(CLASS_WEIGHT) \
                                         --solver $(SOLVER) \
                                         --retrain

$(METRIC_VAL): $(PREDICT_VAL) $(Y_TRN) | $(DIR_METRIC)
	python ./src/evaluate.py --predict-file $< \
                             --target-file $(lastword $^) > $@
	cat $@

$(SUBMISSION_TST): $(PREDICT_TST) $(HEADER) $(ID_TST) | $(DIR_SUB)
	paste -d, $(lastword $^) $< > $@.tmp
	cat $(word 2, $^) $@.tmp > $@
	rm $@.tmp

.DEFAULT_GOAL := all