donalee / taxocom Goto Github PK

Please help !! At first I thought it's the issue of mismatch of terms in terms.txt & seed_taxo. So I manually corrected all and system is able to read but again I got the issue.
####################
Starting cell * with 323 docs.
Starting training using file ../data/auto/root_seed_taxo/docs.txt
Loading embedding from file ../data/auto/root_seed_taxo/embeddings.txt
Category name file: ../data/auto/root_seed_taxo/child_names.txt
Target category charging : charging Smart products EV plugs cables connectors stations Charger
Target category Battery : Battery sustainable swapping models service interoperability
Target category bidirectional : bidirectional charging home charger CCS Smart products EV plugs cables connectors stations Charger
Target category wireless : wireless parking electrified roadways network inductive

[Main] Finish locally discriminative embedding - 3.5880656242370605 seconds
Traceback (most recent call last):
File "main.py", line 71, in
main(args)
File "main.py", line 58, in main
recur(input_corpus, root_dir, '*', args.filter_tau, args.n_locterms, args.betas, 0)
File "main.py", line 23, in recur
child_names = run_clustering(input_corpus, filter_tau, betas, node_name, node_dir, level)
File "/home/fidev/taxocom/code/clustering.py", line 101, in run_clustering
doc_clusters = dataset.get_doc_clusters(term_clusters, term_rel_scores)
File "/home/fidev/taxocom/code/dataset.py", line 185, in get_doc_clusters
term_assignment[term_id] = clus_id
IndexError: list assignment index out of range
Compress the output taxonomy
#################

Hi,
I followed the steps suggested in readme file specific to "data", the code runs well and all the files are generated in the input folder but with very limited or no data.
Please help with detailed steps to create docs.txt & terms.txt as they have been used as an input.
Thanks in advance!! Please help.
Regards,
Gaurav