Update README.md #23
Update README.md #23
Conversation
Updated everything except Docker instructions.
README.md
Outdated
| * X: input data of size (number of patients, number of features). Patients must be sorted with respect to event or censoring times 'T'. | ||
| * T: Time of event or time to last follow-up, appearing in increasing order and corresponding to the rows of 'X'. size: (number of patients, ). | ||
| * O: Right-censoring status. A value of 1 means the event is observed (i.e. deceased or disease progression), a 0 value indicates that the sample is censored. size:(number of patients, ). | ||
| * A: An index array encoding the at-risk set for each sample. For sample _i_, the _i_ th element in A is the index of the next sample in the at risk group of _i_ (see Run.py for an example of how to calculate this vector using the provided functions) size:(number of patients, ). |
There was a problem hiding this comment.
@safooray This is confusing - need to describe this better.
|
|
||
| The Run.py training module takes as input a python dictionary containing the following numpy arrays: | ||
| * X: input data of size (number of patients, number of features). Patients must be sorted with respect to event or censoring times 'T'. | ||
| * T: Time of event or time to last follow-up, appearing in increasing order and corresponding to the rows of 'X'. size: (number of patients, ). |
README.md
Outdated
|
|
||
|
|
||
|
|
||
| The provided example scripts assume the data is a .mat file containinig, 'Survival', 'Censored', and either 'Integ\_X' or 'Gene\_X' depending on what feature set we are using. |
There was a problem hiding this comment.
Need to clarify that this is a one-off for how we wrote the paper. We can add some commentary about using the TCGAIntegrator package to generate data.
Updating Docker text.
README.md
Outdated
|
|
||
|
|
||
| will pull the Docker Image, run the container, and mount the host machine data directory with container data path. | ||
| The container Data Path is ***/SurvivalNet/data/<data_file_name>***. |
There was a problem hiding this comment.
@cramraj8 please take a look at this. It seems clear that users have to mount a folder containing their data, but the details are a little unclear. We should add a bit more detail here.
There was a problem hiding this comment.
I will make some more explanation here.
There was a problem hiding this comment.
I already made a pass at this - please review my version for accuracy.
|
|
||
| where users can see how to train and validate deep survival models. The main script | ||
|
|
||
| python Run.py |
There was a problem hiding this comment.
@cramraj8 Is Run.py in the examples folder? Are there other files in the examples folder?
Clarified mounting process.
simplified user guide
safooray
left a comment
safooray
left a comment
There was a problem hiding this comment.
@cooperlab Please take another look.
Added explicit references to readme.
Pull request for collaborative commenting and editing of documentation.