PaddleRec icon indicating copy to clipboard operation
PaddleRec copied to clipboard

缺少reader.py

Open w5688414 opened this issue 3 years ago • 5 comments

https://github.com/PaddlePaddle/PaddleRec/blob/release/2.1.0/datasets/ali-ccp/data_process.sh

缺少reader.py

echo "preprocessing data......"
python reader.py --train_data_path ${train_target_path} \
                 --test_data_path ${test_target_path} \
                 --vocab_path vocab/vocab_size.txt \
                 --train_sample_size 6400 \
                 --test_sample_size 6400 \

w5688414 avatar Jul 18 '21 02:07 w5688414

https://github.com/PaddlePaddle/models/pull/4531 原始处理可以参考这个pr

frankwhzhang avatar Jul 20 '21 06:07 frankwhzhang

不是很明白,请问reader.py对应哪个文件

w5688414 avatar Jul 20 '21 08:07 w5688414

PaddleRec/multi-task/ESMM/reader.py 这个pr中的

frankwhzhang avatar Aug 03 '21 07:08 frankwhzhang

@frankwhzhang 上述PR似乎 只保留 feature_filed: feature_id, 而具体特征的取值去除了

jxlijunhao avatar Apr 19 '22 02:04 jxlijunhao

@frankwhzhang 上述PR似乎 只保留 feature_filed: feature_id, 而具体特征的取值去除了

同疑问

Helafeng avatar May 13 '22 03:05 Helafeng