4. Parameter files Flashcards

1
Q

How can you add parameter files to your model?

A
  • You can simply drag and drop into the appropriate folder or right-click on the folder and select Upload Files
  • They can be JSON or CSV, although json files are faster.
  • Create a subfolder inside the algorithms folder and name it json_parameter_files. CSV files go into the parameter_tables folder
  • The files are separated like this due to how hx Renew is structured. For a file to be accessed from within rating.py, it must be located inside the ‘algorithms’ folder. Except csv files that are located in the parameter_tables folder, where the ‘hx.params’ function can be used to extract them.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the layout of the json parameter file?

A
  • The JSON files are composed of key:value pairs, similar to how the example data is structured. For example
    {
    “Manufacturing”: 0.16,
    “Distributing”: 0.02
    }
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the layout of the csv parameter file?

A
  • The top row contains your column headers, separated by commas.
  • Again, you cannot have spaces in your column headers and should always use lower case.
  • Your header names will be how you access the data in that column, either in the data schema or in your rating algorithm.
  • Your data for each column then follows underneath. For example:
  • size_gte,adjustment
    0,0.2
    7,14
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How do you access parameter data?

A
  • first extract it from the JSON or CSV files. This allows other files (like view.js and rating.py) to retrieve this data.
  • For the JSON file, we will load them into the data schema using the get_parameters function. Put from algorithms.json_parameter_files.parameters import get_parameters into the data schema.
  • The get_parameters function should sit inside the json_parameter_files folder that you created earlier. It is a hx-built function, and you are able to copy and paste the code below:
    import os
    import json

-Function takes the file name as an argument
def get_parameters(file_name):

# Get the path to the parent directory
parent_dir = os.path.dirname(\_\_file\_\_)

# Get the path to the specific file by joining the parent directory with the file name
file_path = os.path.join(parent_dir, file_name)

# Return the json from the parameter file
with open(file_path) as f:
    parameter_json = json.load(f)
return parameter_json
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How do you define parameters from a csv?

A
  • Open parameter_tables_schema.json
  • “csv_table_name”:{
    “column_name”:”column_type”
    }
How well did you know this?
1
Not at all
2
3
4
5
Perfectly