Download preprocess_data/EXAMPLE_INPUTS/run_clean_data_example.sh from latticetower/nested-tkf-repro-workspace: direct link, hf CLI and curl.
- Browser
- Download file 304 Bytes
-
https://huggingface.co/datasets/latticetower/nested-tkf-repro-workspace/resolve/main/preprocess_data/EXAMPLE_INPUTS/run_clean_data_example.sh
- Command line
-
hf download hf://datasets/latticetower/nested-tkf-repro-workspace/preprocess_data/EXAMPLE_INPUTS/run_clean_data_example.sh
-
curl -L -o run_clean_data_example.sh https://huggingface.co/datasets/latticetower/nested-tkf-repro-workspace/resolve/main/preprocess_data/EXAMPLE_INPUTS/run_clean_data_example.sh
304 Bytes
| python clean_data.py \ | |
| -pfam_seed_file EXAMPLE_INPUTS/EXAMPLE_Pfam-A.seed \ | |
| -tree_dir EXAMPLE_INPUTS/trees \ | |
| -num_splits 2 \ | |
| -metadata_header header \ | |
| -rand_key 42 \ | |
| -topk1_valid 0 \ | |
| -topk2_valid 0 \ | |
| -alphabet_size 20 \ | |
| -max_len 5000 \ | |
| -batch_size 10 | |