Open calebgeniesse opened 5 years ago
I've made a singularity container with dyneusr and jupyter. It seems to load everything fine but does run into some errors in the sample notebooks, maybe because they were with an older version? for example:
# run MAPPER
lens = mapper.fit_transform(X, projection=projection)
graph = mapper.map(lens, X_inverse, clusterer, coverer=cover)
# cache the parameters and results for later
mapped = Bunch(X=X, lens=lens, graph=graph, clusterer=clusterer, cover=cover)
output:
..Composing projection pipeline of length 1:
Projections: [0, 1]
Distance matrices: False
Scalers: MinMaxScaler(copy=True, feature_range=(0, 1))
..Projecting on data shaped (100, 3)
..Projecting data using: [0, 1]
..Scaling with: MinMaxScaler(copy=True, feature_range=(0, 1))
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-29-3e3974f3d0af> in <module>
1 # run MAPPER
2 lens = mapper.fit_transform(X, projection=projection)
----> 3 graph = mapper.map(lens, X_inverse, clusterer, coverer=cover)
4
5 # cache the parameters and results for later
TypeError: map() got an unexpected keyword argument 'coverer'
But overall it seems to work fine.
I've updated the docs to mention how to use this image in a pull request #10
If you link your github account to singularity-hub, the singularity recipe which is built on singularity-hub could live in this repository in a sub folder which might be better than in the repository I have but either way
thanks for setting this up, and for updating the docs!
pretty sure you are right about the notebook errors, they were developed using an older version of kmapper
, but i can look into it and make sure they are compatible with more recent versions
and great idea, just linked my github account, but not sure how to connect this repository... 😬
For sure! I will update the pull request
just merged the PR, too excited to try this out myself
Create docker app to make it easier to install/run dyneusr (e.g. on clusters)
TODO: