feat: add export_statistics and export_mentions make targets with dataset prefix markers

Add two new Makefile targets for exporting statistics and mentions data from the dataset file, each prefixed with "===" markers in the dataset.py dump function to enable clean grep extraction. The export_statistics target filters lines containing "Statistics:" while export_mentions filters lines containing "times ment", both writing results to separate export files.
This commit is contained in:
2024-11-24 06:09:50 +00:00
parent f06382c373
commit 9497cde9c8
2 changed files with 41 additions and 11 deletions
+15 -1
View File
@@ -1,4 +1,4 @@
all: build sync_excempt export_dataset export_stats merge_images
all: build sync_excempt export_dataset export_stats merge_images export_statistics export_mentions
build:
pip install build
@@ -27,6 +27,20 @@ export_dataset:
@echo "Exporting dataset to be used for LLM embedding. Result will be ./export/0_dataset.txt"
dr.dataset > export/0_dataset.txt
export_statistics:
@echo "Exporting statisticts. Result will be ./export/2_statistics.txt"
cat export/dataset.txt | grep "Statistics: "
cat export/dataset.txt | grep "Statistics: " > export/2_statistics.txt
export_mentions:
@echo "Exporting mentions. Result will be ./export/3_mentions.txt"
cat export/dataset.txt | grep "times ment"
cat export/dataset.txt | grep "times ment" > export/3_mentions.txt
merge_images:
@echo "Merging images to one big image. Result will be ./export/1_graphs_compliation.png."
python merge_images.py