examples/github_issue_summarization/Pachyderm_Example/pre_process.json

24 lines
645 B
JSON

{
"pipeline": {
"name": "pre_process"
},
"transform": {
"image": "pachyderm/gh-sum:v7",
"cmd": [
"python",
"./workspace/src/preprocess_data_for_deep_learning.py",
"--input_traindf_csv=/pfs/split/github_issues_medium_train.csv",
"--output_body_preprocessor_dpkl=/pfs/out/body_preprocessor.dpkl",
"--output_title_preprocessor_dpkl=/pfs/out/title_preprocessor.dpkl",
"--output_train_title_vecs_npy=/pfs/out/train_title_vecs.npy",
"--output_train_body_vecs_npy=/pfs/out/train_body_vecs.npy"
]
},
"input": {
"pfs": {
"repo": "split",
"glob": "/"
}
}
}