using pykanto from development mode

[gg4u]

I am trying to circumvent the issue:
#18 (comment)

I reinstalled pykanto as shown here:

git clone https://github.com/nilomr/pykanto.git
cd pykanto
pip install -e '.[dev, test, doc]'

I can bypass the issue above in the calc_chunks() function, /utils/compute.py:

   if not n_workers:
        try:
            n_workers = len(psutil.Process().cpu_affinity())
        except:
            print('Mac OS does not support cpu_affinity() in psutil. Default to 4 workers')
            n_workers = 4

And will follow the basic workflow steps:

from pykanto.utils.paths import pykanto_data
from pykanto.dataset import KantoData
from pykanto.parameters import Parameters


DATASET_ID = "GREAT_TIT"
DIRS = pykanto_data(dataset=DATASET_ID)
# ---------
params = Parameters() # Using default parameters for simplicity, which you should't!
dataset = KantoData(DIRS, parameters=params, overwrite_dataset=True)
dataset.data.head(3)

but now FileNotFoundError is raised:

---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
Cell In[122], line 2
      1 DATASET_ID = "GREAT_TIT"
----> 2 DIRS = pykanto_data(dataset=DATASET_ID)
      3 # ---------
      4 params = Parameters() # Using default parameters for simplicity, which you should't!

File ~/Sites/miniconda3/lib/python3.9/site-packages/pykanto/utils/paths.py:431, in pykanto_data(dataset)

File ~/Sites/miniconda3/lib/python3.9/site-packages/pykanto/utils/paths.py:89, in ProjDirs.__init__(self, PROJECT, RAW_DATA, DATASET_ID, mkdir)
     80 def __init__(
     81     self,
     82     PROJECT: Path,
   (...)
     87 
     88     # Type check input paths
---> 89     d = ValidDirs(PROJECT, RAW_DATA, DATASET_ID)
     91     # Define project directories
     92     self.PROJECT = d.PROJECT

File <attrs generated init pykanto.utils.types.ValidDirs>:6, in __init__(self, PROJECT, RAW_DATA, DATASET_ID)
      4 self.DATASET_ID = DATASET_ID
      5 if _config._run_validators is True:
----> 6     __attr_validator_PROJECT(self, __attr_PROJECT, self.PROJECT)
      7     __attr_validator_RAW_DATA(self, __attr_RAW_DATA, self.RAW_DATA)
      8     __attr_validator_DATASET_ID(self, __attr_DATASET_ID, self.DATASET_ID)

File ~/Sites/miniconda3/lib/python3.9/site-packages/attr/_make.py:2146, in _AndValidator.__call__(self, inst, attr, value)
   2144 def __call__(self, inst, attr, value):
   2145     for v in self._validators:
-> 2146         v(inst, attr, value)

File ~/Sites/miniconda3/lib/python3.9/site-packages/pykanto/utils/types.py:46, in f_exists(instance, attribute, f)
     42 """
     43 File exists validator for attr.s decorator.
     44 """
     45 if not f.exists():
---> 46     raise FileNotFoundError(f)

FileNotFoundError: /Users/gg4u/Sites/miniconda3/lib/python3.9/site-packages/pykanto

Does the above code automatically set the path and download the db ?
Or is me that have to set correct path in advance ?

It seems cannot find the PROJECT directory.|

Thanks for helping Nilo!!

[nilomr]

Thanks for raising this. I can't seem to reproduce the error— @gg4u could you run the following and share the output?

import pkg_resources
from pathlib import Path
pydata = Path(pkg_resources.resource_filename("pykanto", "data"))
print(pydata, pydata.parent.exists())

For reference, within a conda environment (you seem to have installed pykanto on your base env), this is my output:

/home/nilomr/miniconda3/envs/pykanto-testing/lib/python3.9/site-packages/pykanto/data True

Thanks!

[gg4u]

Hi! thanks for following up with me :)

I uninstalled pykanto that was installed via pip, so that I can edit and circumvent errors:
#18 (comment)

I installed pykanto by cloning your repo and installing it from there:
pip install -e '.[dev, test, doc]'

For testing, I am using a notebook in the /pykanto/ folder of the cloned git.

The above code yields:

/Users/[MYPATH]/Sites/pykanto/pykanto/data True