1/19/2017

Apply data to R

Today, I reviewed R codes and started to apply the data to R.

I have sent you the data by email. If there is no problem, I will work on them.

I have done the depth matching. They are now can be applied to R without hesitation.

Tomorrow, I will continue to split the data and pre-process the data.
Posted by at

12 comments:

  1. siddharth misraThursday, January 19, 2017 at 10:41:00 PM CST

    it should be "apply R to the data" not the other way round

    ReplyDelete
  2. siddharth misraThursday, January 19, 2017 at 10:45:00 PM CST

    NMR t2 distribution is a measurement.... NMR permeability is a estimation from the t2 distribution...

    I think NMR t2 distribution should be predicted as it is a direct measurement from the tool... once you develop your predictive code, we can also test this using lab measurement with Dr. Rai and Dr. Carl's group.

    ReplyDelete
  3. siddharth misraThursday, January 19, 2017 at 10:46:00 PM CST

    You cannot predict the entire t2 distribution, so you will have to approximate the t2 with some distribution... either Gaussian or a log normal.

    ReplyDelete
  4. siddharth misraThursday, January 19, 2017 at 10:47:00 PM CST

    Let us follow these steps ....


    1. identify the depths where NMR T2 distribution is present. identify top and bottom zones.

    2. divide the entire NMR depth range into testing and training sets based on the best method of partitioning.

    3. select all other relevant log data (to be used to develop the predictive model) from the depths belonging to the training set.

    4. use the other relevant log data to build the predictive model of NMR T2 distribution. You cannot predict the entire NMR T2 distribution. therefore you need to approximate the NMR t2 distribution to sum of 2 or 3 Gaussians or lognormal distribution depending on the data set.

    5. Find out the total number of unknowns that define the predictive model of NMR t2 distribution

    6. Ensure the number of log data to be used as input to build the predictive model exceeds the unknowns.

    ReplyDelete
    Replies
    1. UnknownFriday, January 20, 2017 at 4:42:00 PM CST

      I now know the meaning of every step. I will try to realize them one by one. Thank you so much for your advice.

      Delete
  5. siddharth misraThursday, January 19, 2017 at 10:48:00 PM CST

    so what is the answer for step 1.

    ReplyDelete
    Replies
    1. UnknownFriday, January 20, 2017 at 1:18:00 PM CST

      8996.5-11124 ft

      Delete
    2. UnknownFriday, January 20, 2017 at 1:32:00 PM CST

      I think randomly dividing method will be fine.

      Delete
    3. UnknownFriday, January 20, 2017 at 1:35:00 PM CST

      GR EDTC
      Wire:DPHZ
      Wire:NPOR
      SonicScanner:AT10
      SonicScanner:AT60
      SonicScanner:AT90
      SonicScanner:DTCO
      SonicScanner:DTSM



      Delete
  6. siddharth misraFriday, January 20, 2017 at 12:10:00 PM CST

    When you report in blog follow the format

    Tasks performed

    Tasks for tomorrow

    Details of task performed today:

    ReplyDelete

Subscribe to: Post Comments (Atom)