File size: 1,280 Bytes
3124aa4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 |
set -e
# generate text2sql training dataset with noise_rate 0.2
python text2sql_data_generator.py \
--input_dataset_path "./data/preprocessed_data/preprocessed_train_spider_amr.json" \
--output_dataset_path "./data/preprocessed_data/resdsql_train_spider_amr.json" \
--topk_table_num 4 \
--topk_column_num 5 \
--mode "train" \
--noise_rate 0.2 \
--use_contents \
--add_fk_info \
--output_skeleton \
--target_type "sql"
# predict probability for each schema item in the eval set
python schema_item_classifier.py \
--batch_size 32 \
--device "0" \
--seed 42 \
--save_path "./models/text2sql_schema_item_classifier_semantic" \
--dev_filepath "./data/preprocessed_data/preprocessed_dev_amr.json" \
--output_filepath "./data/preprocessed_data/dev_with_probs_amr.json" \
--use_contents \
--add_fk_info \
--mode "eval"
# generate text2sql development dataset
python text2sql_data_generator.py \
--input_dataset_path "./data/preprocessed_data/dev_with_probs_amr.json" \
--output_dataset_path "./data/preprocessed_data/resdsql_dev_amr.json" \
--topk_table_num 4 \
--topk_column_num 5 \
--mode "eval" \
--use_contents \
--add_fk_info \
--output_skeleton \
--target_type "sql"
|