Merge pull request #7 from alanmarazzi/dev

Finalize docs & tests

.gitignore

@@ -19,4 +19,5 @@ examples/.ipynb_checkpoints
 *.log
 .classpath
 .project
-.settings
+.settings
+resources/docs.md

README.md

@@ -12,68 +12,25 @@ Dataframes in Clojure. Through [pandas](https://github.com/pandas-dev/pandas). O
 
 ## Disclaimer
 
-This is very alpha, things will change fast, will break and the API is neither complete, nor settled. Since a few people have started playing with this there's a Clojars project available. Please give feedback if you're using this, every kind of contribution is appreciated (for more info check the [Contributing](#contributing) section). At the moment everything is mostly undocumented and untested, I'm currently adding them.
+This is alpha, things will change fast, will break and the API is neither complete, nor settled. Since a few people have started playing with this there's a Clojars project available. Please give feedback if you're using this, every kind of contribution is appreciated (for more info check the [Contributing](#contributing) section). At the moment everything is mostly undocumented and untested, I'm currently adding them.
 
 [![Clojars Project](https://img.shields.io/clojars/v/panthera.svg)](https://clojars.org/panthera)
 
 ## Get started
 
 **Panthera** uses the great [libpython-clj](https://github.com/cnuernber/libpython-clj) as a backend to access Python and get [pandas](https://github.com/pandas-dev/pandas) and [numpy](https://github.com/numpy/numpy) functionality.
 
-### System level
-
-If you usually don't develop in Python then a system level install might be a good solution (though always discouraged), if this is your case then follow the subsequent steps.
-
-To get started you need python, pandas and numpy (the latter comes with the former) on your path. Usually a:
-
-```bash
-sudo apt install libpython3.6-dev
-pip3 install numpy pandas xlrd # the latter is for Excel files, if you don't care you can do without
-```
-
-### Environments
-
-If you want to have different Python environments, then getting **panthera** to work correctly is a bit more tricky.
-
-First create your new environment with at least python=3.6, numpy and pandas. (This was tested both on GNU/Linux and WSL with [conda](https://docs.conda.io/projects/conda/en/latest/), but there's no reason why it shouldn't work with other env management tools. On other systems, [Docker is your best bet](https://github.com/scicloj/docker-hub/tree/master/panthera)):
-
-```bash
-conda create -n panthera python=3.6 numpy pandas
-```
-
-Then check the path to the newly created environment:
-
-```bash
-conda activate panthera
-which python
-```
-
-Now you just have to add to one of your profiles the path to the wanted python executable:
-
-```bash
-{:dev {:resource-paths ["/home/user/miniconda3/envs/panthera"]}}
-```
-
-You can create different profiles with different paths according to what you need. Now if you want to make it possible to work with **panthera** without having to activate your environments you have 2 choices:
-
-- assign `PYTHONHOME` env variable to your environment
-
-```bash
-PYTHONHOME="/home/user/miniconda3/envs/panthera" lein whatever
-```
-
-- assign `PYTHONHOME` env variable before requiring **panthera**
-
-```bash
-(System/setProperty "PYTHONHOME" "/home/user/miniconda3/envs/panthera")
-```
+### N.B.: check [libpython-clj](https://github.com/cnuernber/libpython-clj) repo on how to install and start a Clojure/Python session.
 
 ### The actual code
 
 After this you can start playing around with **panthera**
 
 ```clojure
-(require '[panthera.panthera :as pt])
+(require '[[panthera.panthera :as pt]
+           [libpython-clj.python :refer [initialize!]])
+
+(initialize!)
 
 (-> (pt/read-csv "mycsv.csv")
     (pt/subset-cols "Col1" "Col2" "Col3")
@@ -86,33 +43,7 @@ The above chain will read your csv file as a DataFrame, select only the given co
 
 ## Numpy
 
-All of Numpy is wrapped and accessible through a single interface from `panthera.numpy`.
-
-```clojure
-(require '[panthera.numpy :refer [npy doc]])
-
-(npy :power {:args [[1 2 3] 3]})
-;=> [1 8 27]
-
-(npy :power)
-; This arity returns the actual numpy object that can be passed around to other functions as an argument
-```
-
-To access functions inside submodules pass to `npy` a sequence of keys leading to the wanted function:
-
-```clojure
-(npy [:linalg :svd] {:args [[1 2 3] [4 5 6]]})
-```
-
-You can check the original docstring for every module and function with the `doc` helper
-
-```clojure
-(doc :power)
-
-(doc [:linalg :eigh])
-```
-
-To see what is available and how everything works check the [official docs](https://docs.scipy.org/doc/numpy/reference/) online.
+All of Numpy is accessible through [libpython-clj](https://github.com/cnuernber/libpython-clj) interop, check the repo for more info.
 
 ## Contributing
 
@@ -142,8 +73,8 @@ Panthera doesn't pretend to be a clever wordplay because it doesn't need to. Fir
 
 ## License
 
-Copyright © 2019 Alan Marazzi
+Copyright © 2020 Alan Marazzi
 
 This program and the accompanying materials are made available under the
 terms of the Eclipse Public License 2.0 which is available at
-http://www.eclipse.org/legal/epl-2.0.
+http://www.eclipse.org/legal/epl-2.0.

examples/panthera-intro.ipynb

@@ -71,20 +71,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "nil"
-      ]
-     },
-     "execution_count": 2,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "(require '[panthera.panthera :as pt])"
    ]
@@ -3913,9 +3902,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Clojure (clojupyter-v0.2.2)",
+   "display_name": "Lein-Clojure",
    "language": "clojure",
-   "name": "clojupyter"
+   "name": "lein-clojure"
   },
   "language_info": {
    "file_extension": ".clj",

project.clj

@@ -1,11 +1,11 @@
-(defproject panthera "0.1-alpha.16"
-  :description "Data Frames in Clojure (with Pandas) + NumPy"
-  :url "https://github.com/alanmarazzi/panthera"
-  :scm {:name "git" :url "https://github.com/alanmarazzi/panthera"}
-  :license {:name "EPL-2.0"
-            :url  "https://www.eclipse.org/legal/epl-2.0/"}
-  :dependencies [[cnuernber/libpython-clj "1.0"]
-                 [org.clojure/core.memoize "0.7.2"]]
-  :profiles {:dev {:dependencies [[org.clojure/clojure "1.10.1"]]
-                   :plugins [[refactor-nrepl "2.4.0"]
-                             [cider/cider-nrepl "0.18.0"]]}})
+(defproject panthera "0.1-alpha.17"
+  :description "Data Frames in Clojure (with Pandas) + NumPy"
+  :url "https://github.com/alanmarazzi/panthera"
+  :scm {:name "git" :url "https://github.com/alanmarazzi/panthera"}
+  :license {:name "EPL-2.0"
+            :url  "https://www.eclipse.org/legal/epl-2.0/"}
+  :dependencies [[cnuernber/libpython-clj "1.32"]
+                 [org.clojure/core.memoize "0.7.2"]]
+  :profiles {:dev {:dependencies [[org.clojure/clojure "1.10.1"]]
+                   :plugins [[lein-jupyter "0.1.16"]]}})
+