invisum-api

API for the In Visum service demo.

Install

We are using the Continuum Analytics Anaconda platform for the necessary data science tools. You can try aliasing all anaconda stuff to get easier access:

$ cd ~/anaconda2/bin
$ for i in *; do alias "ana-$i"="$(pwd -P)/$i"; done

Install depedencies with:

$ pip install -r requirements.txt

To setup the first time:

$ ana-python manage.py makemigrations datasets
$ ana-python manage.py migrate

To run:

$ ana-python manage.py runserver

Configuration

Please set a SECRET_KEY in your environment variables before executing. If you need to generate a new key, you can use online tools like this.

---

API Documentation

Permissions

• AllowAny: Unrestricted access - default
• IsAuthenticatedOrReadOnly: Need to be authenticated for non safe methods.
• IsAuthenticated: Need to be authenticated for any method.
• IsOwnerOrReadOnly: Only owners of a object can edit.

Requests

• datasets/
  • Name: dataset-list
  • Permissions: IsAuthenticatedOrReadOnly
  • GET: List all datasets.
  • POST: Create new dataset.
• datasets/<id>/
  • Name: dataset-detail
  • Permissions: IsAuthenticatedOrReadOnly, IsOwnerOrReadOnly
  • GET: Information about <id> dataset.
  • PUT/PATCH: Update <id> dataset.
  • DELETE: Delete the <id> dataset.
• datasets/personal/<id>/
  • Name: personal-create
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly
  • GET: List all personal datasets related to <id> and the user.
  • POST: Create a new personal dataset related to <id> and the user.
• personal/
  • Name: personal-list
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly
  • GET: List all personal datasets related to the user.
• personal/<id>/
  • Name: personal-detail
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly
  • GET: Information about <id> personal dataset.
  • PUT/PATCH: Update <id> personal dataset.
  • DELETE: Delete the <id> personal dataset.
• users/
  • Name: user-list
  • GET: List all users.
• users/<id>/
  • Name: user-detail
  • GET: Information about <id> user.
• datasets/rate/<id>/
  • Name: rating-list
  • Permissions: IsAuthenticated
  • GET: Retrieve the logged user rating related to <id> dataset.
  • POST: Create new rating for <id> dataset(Only one per dataset-user).
• ratings/<id>/
  • Name: rating-detail
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly
  • GET: Retrieve the <id> rating.
  • PUT/PATCH: Update the <id> rating.
  • DELETE: Delete the <id> rating.
• search/users/<text>/
  • Name: search-users
  • GET: Search datasets using <text> by owner username.
• search/title/<text>/
  • Name: search-title
  • GET: Search datasets using <text> by title.
• discover/
  • Name: discover-feed
  • Permissions: IsAuthenticated
  • GET: List the 10 most ranked datasets.
• personal/operation/<op>/<id>/
  • Name: dataset-operation
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • POST: Execute a new operation <op> in personal dataset <id>.
• personal/operation/<op>/<l-id>-<r-id>/
  • Name: multiset-operation
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • POST: Execute a new multiset operation <op> in personals datasets <l-id> and <r-id>.
• personal/plot/<chart>/<id>/
  • Name: plot-create
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • POST: Create a new chart <chart> based on dataset <id>.
• personal/plot/
  • Name: plot-list
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • GET: List the plots related to the user.
• personal/<id>/meta/
  • Name: personal-meta
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • GET: Get information about the personal dataset <id>.
• personal/plot/<id>/
  • Name: plot-serve
  • Permissions: IsAuthenticated, IsOwnerOrReadOnly*
  • GET: Serve the plot <id> to the user.
  • DELETE: Delete the plot <id>.

Single Dataset Operations

1. Slice
   • Simple pythonic slice.
   • Arguments: left, right, step
2. Drop
   • Drop columns/rows based on labels.
   • Arguments: labels, axis
   • Reference
3. Filter
   • Subset of columns/rows based only on labels.
   • Arguments: items, like, regex, axis
   • Reference
4. Fillna
   • Fill non existent values.
   • Arguments: value
   • Reference
5. Dropna
   • Drop columns/rows with non existent values.
   • Arguments: how
   • Reference
6. Sort
   • Sort values.
   • Arguments: by, ascending, axis, na_position
   • Reference

Multi Dataset Operations

1. Merge
   • Merge two personal datasets.
   • Arguments: how, on, left_on, right_on, right_index, sort, suffixes, copy, indicator
   • Reference
2. Closest Match Merge (Testing)
   • Merge two personal datasets based on the closest match.
   • Arguments: left_pivot, right_pivot, how, on, left_on, right_on, right_index, sort, suffixes, copy, indicator
   • Merge Reference
   • Closest Match Reference

Charts

Reference

Base configuration

• plot_width
• plot_height
• legend

Chart Configuration

1. Histogram
   • Arguments: values, label, agg, bins, density
2. Bar
   • Arguments: values, label
3. Line
   • Arguments: x, y
4. Scatter
   • Arguments: x, y