Merge pull request #2 from CryoInTheCloud/fperez-packages

Add packages for everyday usage

environment.yml

@@ -1,51 +1,107 @@
+name: cryointhecloud
+
 channels:
   - conda-forge
 
 dependencies:
-  - python=3.10
+  - python==3.10.8
   # Required until https://github.com/jupyterhub/repo2docker/pull/1196 is merged
   - jupyterhub-singleuser>=3.0,<4.0
   # nbgitpuller is very helpful when distributing user content
-  - nbgitpuller
+  - nbgitpuller==1.1.1
   # Specify major version of jupyterlab to use. Manually bump it when you want to upgrade
   # jupyterlab version
   - jupyterlab>=3.0,<4.0
   # This provides memory usage stats on JupyterLab and classic notebook
-  - jupyter-resource-usage
+  - jupyter-resource-usage==0.6.4
   # https://syncthing.net/ provides a dropbox-like interface for syncing directories across users,
   # hubs and machines. Used along with `jupyter-syncthing-proxy` from pip.
-  - syncthing
+  - syncthing==1.22.1
   # gh-scoped-creds allows users to securely push to GitHub from their repo.
   # https://blog.jupyter.org/securely-pushing-to-github-from-a-jupyterhub-3ee42dfdc54f has
   # some more information
-  - gh-scoped-creds
+  - gh-scoped-creds==4.1
 
   # Cloud access tools and libraries
   # Access S3 easily from python
-  - s3fs
+  - s3fs==2022.11.0
   # handle login and similar details for accessing earthdata protected data
   # https://github.com/nsidc/earthaccess
-  - earthdata
+  - earthdata==0.4.1
   # Access datasets exposed via STAC
-  - pystac-client
+  - pystac-client==0.5.1
+  # Access datasets exposed via intake catalogs
+  - intake==0.6.6
+  - intake-esm==2022.9.18
+  - intake-xarray==0.6.1
+
+  # Core scientific python stack
+  - numpy==1.23.5
+  - scipy==1.9.3
+  - matplotlib==3.6.2
+  - pandas==1.5.2
+  - networkx==2.8.8
+  - scikit-image==0.19.3
+  - scikit-learn==1.1.3
+  - sympy==1.11.1
+  - statsmodels==0.13.5
+  - xarray==2022.11.0
+  - hypothesis==6.58.0 # needed by numpy testing tools
+  - cython==0.29.32 # optimization, C API access
+  - numba==0.56.4 # high-performance numerics
+
+  # Jupyter tools
+  - ipython==8.6.0
+  - ipywidgets==7.7.2
+  - jupyterlab-favorites==3.1.0
+  - jupyterlab-geojson==3.3.1
+  - jupyterlab-git==0.39.3
+  - jupyter-book==0.13.1
+  - nbconvert==6.5.3
+  - nbdime==3.1.1
 
   # Visualization packages
-  - matplotlib
-  - ipympl # This enables matplotlib interaction with jupyter widgets
-  - ipyleaflet
+  - cartopy==0.20.2 # geospatial plotting with matplotlib
+  - seaborn==0.12.1 # statistical plotting with matplotlib
+  - ipympl==0.9.2 # This enables matplotlib interaction with jupyter widgets
+  - ipyleaflet==0.17.2
+  - bokeh==3.0.2
+  - plotly==5.11.0
 
   # Geospatial data packages
-  - rasterio
-  - rioxarray
-  - xarray-datatree
+  - geopandas==0.12.1
+  - rasterio==1.2.10
+  - rioxarray==0.13.1
+  - xarray-datatree==0.0.9
+
+  # File formats and file management, download, dataset caching
+  - h5py==3.7.0
+  - hdf5==1.12.1
+  - h5netcdf==1.1.0
+  - pooch==1.6.0
+
+  # Distributed computing
+  - dask==2022.11.0
+  - dask-labextension==6.0.0
+
+  # Other useful generic python packages
+  - pillow==9.2.0 # Python imaging library, useful for many image-related tasks
+  - tqdm==4.64.1 # progress bars, with notebook support
+  - pytest==7.2.0
+  - pytest-cov==4.0.0
+  - pep8==1.7.1
+  - flake8==5.0.4
 
   # Packages specific to cryo work
-  - icepyx
+  - icepyx==0.6.4
+
+  # Desktop tools whose versions are more recent on conda-forge than ubuntu
+  - qgis==3.18.3
 
   # Not all packages will be available from conda-forge, we install from pip when we need to.
-  - pip
+  - pip==22.3.1
   - pip:
-      - jupyter-syncthing-proxy
+    - jupyter-syncthing-proxy
 
       # Access linux desktop from inside JupyterHub
-      - jupyter-desktop-server
+    - jupyter-desktop-server