llano1025 / desc2mdf Goto Github PK

This project use AI to turn description to MDF group and recommended form.

1. Create virtual env. The 1st venv is option. The 2nd venv is we named the virtual env as ‘venv’

   python -m venv venv

2. Active the virtual venv by gitbash

   source venv/Scripts/activate

3. Install all dependency

   pip install -r requirements.txt

4. Data training is necessary in first run. Update following configable.
   IS_TRAIN_MODEL_NEEDED = True
   
5. Python run main.py

1. Active the virtual venv by gitbash

   source venv/Scripts/activate

2. Install all dependency

   pip install -r requirements.txt

3. Data training is optional. Update following configable.
   IS_TRAIN_MODEL_NEEDED = False
   
4. Python run main.py

• Training data: data\BasicInfo_Result.xlsx
• Input data: data\beseformDB.mdfAssetDescriptions.xlsx

• Training data: data\recForm.xlsx

• Input data: data\beseformDB.mdfAssetDescriptions.xlsx

• No. of epochs: self.epochs = 50 (default)

• Output data: exc/outputs/MDF2Rec_output.xlsx

• Output data: exc/outputs/Desc2MDF_output.xlsx

Start the flask local server

flask --app main run

Example request

curl --location 'http://localhost:5000'
--header 'Content-Type: application/json'
--data '{"desc": "['''ABC''', '''IMPULSION PUMP, LONG''', '''POWER SUPPLIER''']"}'

Example reponse

{
  "data": [
    {
      "desc": "ABC",
      "groupNo": "mm-pm-00",
      "recommendedForm": "MDF15"
    },
    {
      "desc": "IMPULSION PUMP, LONG",
      "groupNo": "mm-pm-32",
      "recommendedForm": "MDF01"
    },
    {
      "desc": "POWER SUPPLIER",
      "groupNo": "mm-pm-01",
      "recommendedForm": "MDF15"
    }
  ]
}

1. Create virtual env. The 1st venv is option. The 2nd venv is we named the virtual env as ‘venv’

   python -m venv venv

2. Active the virtual venv by gitbash

   source venv/Scripts/activate

3. Install all dependency

   pip install -r requirements.txt

4. Start server

   flask --app main run