The issue with the custom_attr_name_dict parameter in the tokenizer.py
#82
by
mriee
- opened
In the tokenizer.py, the custom_attr_name_dict parameter has keys that represent the names of the attributes in the loom file, and values that represent the names of the attributes in the dataset. However, the program only uses the keys and does not use the values. In other words, based on my understanding, the program should be as follows:
loom_cell_attr = [attr_key for attr_key in custom_attr_name_dict.keys()]
cell_metadata = {attr_key: [] for attr_key in custom_attr_name_dict.values()}
...
for k in loom_cell_attr:
cell_metadata[custom_attr_name_dict[k]] += file_cell_metadata[k]
Am I understanding this correctly? Thank you!
I also faced the same problem when I used tokenizer.py.
In the dataset. file only has the key not the value.
Thank you for noticing this and suggesting a solution! I updated the tokenizer accordingly.
ctheodoris
changed discussion status to
closed