Code snippet in README throws an error

[yamsgithub]

I am running the code from the README but it throws an error:

---

from dsbox.datapreprocessing.profiler import Profiler
import pandas as pd

profiler = Profiler()
data = pd.read_csv('test.csv', dtype=object)
jsonResult = profiler.produce(inputs=data)

ImportError Traceback (most recent call last)
in ()
----> 1 from dsbox.datapreprocessing.profiler import Profiler
2 import pandas as pd
3
4 profiler = Profiler()
5 data = pd.read_csv('/Users/yamuna/D3M/data/185_baseball/185_baseball_dataset/tables/learningData.csv', dtype=object)

ImportError: cannot import name Profiler

[kyao]

It looks like Python cannot find the Profiler.

We need to update our readme file.

To install our profiler, please clone the master branch. Then, cd to your local repo directory, and do:

pip install -e .

Also, for sample usage you can take a look at:

https://github.com/usc-isi-i2/dsbox-profiling/blob/master/ta1-pipeline.py

[yamsgithub]

I am running the ta1-pipeline.py but gettting the following error. Where are the config files?

FileNotFoundError Traceback (most recent call last)
in ()
19
20 # Load the json configuration file
---> 21 with open("ta1-pipeline-config.json", 'r') as inputFile:
22 jsonCall = json.load(inputFile)
23 inputFile.close()

FileNotFoundError: [Errno 2] No such file or directory: 'ta1-pipeline-config.json'

[kyao]

NIST provides that file when we submit our pipelines for testing. That file looks

{ "train_data": "/path-to-data/seed_datasets_current/38_sick/TRAIN", "test_data": "/path-to-data/seed_datasets_current/38_sick/TEST", "output_folder": "." }

[yamsgithub]

Thanks. I was able to run it. I am using the following config.json:
{ "train_data": "/Users/yamuna/D3M/data/196_autoMpg/TRAIN", "test_data": "/Users/yamuna/D3M/data/196_autoMpg/TEST", "output_folder": "." }

But now get this error:

python ta1-pipeline.py > profiling_output.log
Traceback (most recent call last):
File "ta1-pipeline.py", line 57, in
ds2.metadata.pretty_print()
File "/Users/yamuna/D3M/ta2/src/d3m-metadata/d3m_metadata/metadata.py", line 646, in pretty_print
self.pretty_print(selector + [element], handle=handle, _level=_level + 1)
File "/Users/yamuna/D3M/ta2/src/d3m-metadata/d3m_metadata/metadata.py", line 635, in pretty_print
self.pretty_print(selector + [ALL_ELEMENTS], handle=handle, _level=_level + 1)
File "/Users/yamuna/D3M/ta2/src/d3m-metadata/d3m_metadata/metadata.py", line 646, in pretty_print
self.pretty_print(selector + [element], handle=handle, _level=_level + 1)
File "/Users/yamuna/D3M/ta2/src/d3m-metadata/d3m_metadata/metadata.py", line 625, in pretty_print
for line in json.dumps(query(selector=selector), indent=1, cls=MetadataJsonEncoder).splitlines():
File "/Users/yamuna/anaconda3/lib/python3.6/json/init.py", line 238, in dumps
**kw).encode(obj)
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 201, in encode
chunks = list(chunks)
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 438, in _iterencode
yield from _iterencode(o, _current_indent_level)
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 430, in _iterencode
yield from _iterencode_dict(o, _current_indent_level)
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 404, in _iterencode_dict
yield from chunks
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 437, in _iterencode
o = _default(o)
File "/Users/yamuna/D3M/ta2/src/d3m-metadata/d3m_metadata/metadata.py", line 182, in default
return super().default(o)
File "/Users/yamuna/anaconda3/lib/python3.6/json/encoder.py", line 180, in default
o.class.name)
TypeError: Object of type 'int64' is not JSON serializable

[kyao]

We submitted a fix to this bug. Are you using an older version of d3m metadata? Try using the version with tag v2018.1.26