Deep Learning with TensorFlow | Nissim

Discussion in 'Big Data and Analytics' started by Nishant_Singh, Jul 14, 2019.

  1. Nishant_Singh

    Nishant_Singh Well-Known Member
    Simplilearn Support

    Joined:
    Aug 1, 2018
    Messages:
    251
    Likes Received:
    55
    #1
    _34747 and Shrinivas M Bakale like this.
  2. Nirmal Kumar P M

    Joined:
    May 15, 2019
    Messages:
    5
    Likes Received:
    0
    What we were taught is to use ANNs for classification data. Can it be used to train models of a continuous label data? If yes, could you provide some inputs how to use the ANN in this case?
     
    #2
  3. Shrinivas M Bakale

    Alumni

    Joined:
    Apr 13, 2016
    Messages:
    3
    Likes Received:
    0
    Y
    Yes, we can. But, no Activation function required due to Continuous output
     
    #3
  4. Nirmal Kumar P M

    Joined:
    May 15, 2019
    Messages:
    5
    Likes Received:
    0
    Is it possible to use the hyper parameter tuning method GridsearchCV with linear regression models and ANNs?
     
    #4
  5. Dilanjan

    Dilanjan Member

    Joined:
    Apr 9, 2019
    Messages:
    5
    Likes Received:
    0
    Hi Nissim
    I tried to work on the project "Deep Learning with Keras and Tensorflow" - "Lending Club Loan Data Analysis" and downloaded
    the datasets it has 3 files, loan_data has the feature names (14 columns) but other 2 files has no column names one has 18 columns other file has two columns, could you please explain the 3 dataset files

    Regadrs
    Dilanjan
     
    #5
  6. Vikas Kumar_18

    Vikas Kumar_18 Well-Known Member
    Simplilearn Support Alumni

    Joined:
    Dec 17, 2018
    Messages:
    205
    Likes Received:
    35
    Hi Dilanjan,

    Greetings from Simplilearn!

    1. loan_data : It is the mother data but no need to use this data.

    2. input : You need to use this data and this data does not have the column names. Purposefully it is removed. When real-time data is collected then prevent the data breach, actual column name is removed so that no one can know about the real entity. You have to understand this data as "X" and should split it as (X_train, X_test).

    3. output: This is label of your input data. You need to consider this complete two columns as "Y" while splitting. It has two columns so don't worry about this. You should split this Y data as (Y_train and Y_test).

    Here you need to use input and output data and split this data into train and test and the play with this

    If you still have confusion then attend the second last and the last day live session where faculty would make you understand the flow of the project.
     
    #6

Share This Page