Trying to save an h2o model with some specific name that differs from the model's model_id field, but trying something like...
h2o.save_model(model=model,
path='/some/path/then/filename',
force=False)
just creates a dir/file structure like
some
|__path
|__then
|__filename
|__<model_id>
as opposed to
some
|__path
|__then
|__filename
Is this possible to do from the save_model method?
I can't / hesitate to simply change the model_id before calling the save method because the model names have timestamps appended to them to avoid name collisions with other models that may be on the h2o cluster (am trying to remove these timestamps when saving on disk and simplifying the name on the cluster before saving creates a time where naming collision can occur if other processes are also attempting to save such a model (of, say, a different timestamp)).
Any way to get this behavior or other common alternatives / workarounds?
This is currently not possible, however I created a feature request here. There is a related question here which shows a solution for R (could be adapted to Python). The work-around is just to rename the file manually using a few lines of R/Python code.
Related
This is my situation. I have over 400 features, many of which are probably useless and often zero. I would like to be able to:
train an model with a subset of those features
query that model for the features actually used to build that model
build a H2OFrame containing just those features (I get a sparse list of non-zero values for each row I want to predict.)
pass this newly constructed frame to H2OModel.predict() to get a prediction
I am pretty sure what found is unsupported but works for now (v 3.13.0.341). Is there a more robust/supported way of doing this?
model._model_json['output']['names']
The response variable appears to be the last item in this list.
In a similar vein, it would be nice to have a supported way of finding out which H2O version that the model was built under. I cannot find the version number in the json.
If you want to know which feature columns the model used after you have built a model you can do the following in python:
my_training_frame = your_model.actual_params['training_frame']
which will return some frame id
and then you can do
col_used = h2o.get_frame(my_training_frame)
col_used
EDITED (after comment was posted)
To get the columns use:
col_used.columns
Also, a quick way to check the version of a saved binary model is to try and load it into h2o, if it loads it is the same version of h2o, if it isn't you will get a warning.
you can also open the saved model file, the first line will list the version of H2O used to create it.
For a model saved as a mojo you can look at the model.ini file. It will list the version of H2O.
Is there a way to search for a resource by its referencing resources? For example, is the a way to find all Observations of code = X with Provenance by agent Y?
GET [base]/Observation?code=X&???
One could:
GET [base]/Provenance?userid=Y&_include=Provenance:target:Observation
but that prevents any kind of filtering on Observation (which may create a volume problem in the response!). Also, I don't need the provenance resource - I just need to make sure that the Observations I'm using have a certain provenance.
Right now, to the best of my knowledge, there's no way to apply filters to multiple resources unless you're using _filter or using a custom OperationDefinition.
This is really more of a "using what method" than a "how-to" question. I am creating a site in NodeJS with Express. So, each user has the ability to upload a profile picture, and my concern is how to route requests for these images. A couple of questions I have are:
Can I use express.static() to serve a default file if a valid one isn't specified? If not, am I going to have to specify a GET route for /img/profileand handle the FS querying there?
How can I find the correct image if multiple file extensions are allowed? (I originally just removed the file extension and they appeared in img tags anyway, is that okay?)
I am currently naming all pictures after their user's name. Looking ahead into the future (for apps I may have to scale), what are normal naming conventions for static user content? Are most stored with a UUID referencing the content in the database?
What else should I take into consideration that I may not have accounted for yet?
First question:
At present, I'd recommend storing your images in a predictable location that can be inferred from some combination of columns or fields in your database entries. One of those fields or columns would be a filename (accounts for different extensions). Then, if they haven't uploaded an image, you just lay down in your view code a reference to the generic "has not set an image" file.
express.static obviously can server static files, but it currently doesn't have a way to serve some other file if the one you wanted isn't there. Because this sounded like fun, I made some modifications to static to support what you request and submitted a couple of pull requests (the feature touched 2 different modules):
In send: https://github.com/visionmedia/send/pull/33
In connect: https://github.com/senchalabs/connect/pull/999
I don't know if those will get included in the project, but if you want to see my forks that have these changes, they are here:
https://github.com/bigohstudios/send
https://github.com/bigohstudios/connect
If this went through, you would mount a static at the root of your profile image files:
app.use(static('<profile image files root>', { fallback: 'anon.jpg'}))
Second question
I generally store the extension in my DB, then when I load the image, I have the correct extension to put into the src attribute on the img tag.
Third question
If you can guarantee unique names, then using those would work. I prefer using the DB id or a UUID as you suggest. It's less intuitive when scanning all the image uploads, but I rarely find myself doing that. I'm usually hunting for a specific image, and I can look up the identifier for that as needed.
Fourth question
You may end up storing the static content outside your app server (e.g. on S3). If that happens, then of course static won't help you.
I try to add AutoSave support to the Core Data File Wrapper example
Now if i have a new/untitled document writeSafelyToURL is called with the NSAutosaveElsewhereOperation type.
The bad thing is, I get this type in both typical use cases
- new file: which store a complete new document by creating the file wrapper and the persistent store file
- save diff: where the file wrapper already exists and only an update is required.
Does somebody else already handled this topic or did somebody already migrated this?
The original sample use the originalStoreURL to distinguish those two use cases, which solution worked best for you?
Thanks
I just created a GUI using guide in MATLAB for a small project I'm working on. I have amongst other things two text fields for from and to dates. Now I'd like to get rid of them and use a Java date select tool. Of course this is not possible using guide so I need to add them manually.
I've managed to get them to show up by putting this code into my Opening_Fcn,
uicomponent(handles, 'style','com.jidesoft.combobox.DateChooserPanel','tag','til2');
using UICOMPONENT.
But even though it shows up I can't access the date select's attributes, for example
get(handles.til2)
returns
??? Reference to non-existent field 'til2'.
How can I fix this?
Unless you edit the saved GUI figure, the basic handles structure will not include your new component by default.
One way to access you component is to store the handle via guidata, by adding the following to your opening function:
handles.til2 = uicomponent(handles, 'style','com.jidesoft.combobox.DateChooserPanel','tag','til2');
guidata(hObject,handles)
Functions that need to access the handle need the line
handles = guidata(hObject)
to return the full handles structure that includes the filed til2