Add soiling notebooks to docs (#51)

* Base_url should not appear in url params when provided

* Add response type hint to live soiling hsu method

* Add HSU notebook example to docs

* Add Kimber notebook example to docs

* unused import

* pyproject cleanup

* soiling notebook updates

* fix precip accum norm bug

* dict | bug

* no pip install in notebooks, use [all], clarify pandas dep

* fix response docs, add jupyter notebook downloads

* forgot to add new files to previous commit

* move dust soiling notebooks from 2. to 1.

* fix soiling notebook titles

* Rerun notebooks to update output cells

---------

Co-authored-by: Will Holmgren <William.Holmgren@dnv.com>

Author: GF-D6

README.md

@@ -24,23 +24,33 @@ cd solcast-api-python-sdk
 pip install .
 ```
 
-The vanilla version doesn't have any dependency. For full functionality,
-for example for getting the data into `DataFrames`, and for development, use the `[all]` tag:
+The base solcast sdk install requires only the python standard library.
+Pandas is the only optional dependency that adds functionality to the package.
 
 ```commandline
-pip install .[all] for the dev libs
+pip install solcast pandas
 ```
 
+The example notebooks use a variety of optional dependencies to showcase different
+ways in which the Solcast API may be used. To install these dependencies run
+
+```commandline
+pip install solcast[all]
+```
+
+
 ## Basic Usage
 
 ```python
 from solcast import live
 
-df = live.radiation_and_weather(
+res = live.radiation_and_weather(
     latitude=-33.856784,
     longitude=151.215297,
     output_parameters=['air_temp', 'dni', 'ghi']
-).to_pandas()
+)
+res.to_dict()
+res.to_pandas()  # requires optional pandas installation
 ```
 
 Don't forget to set your [account Api Key](https://toolkit.solcast.com.au/register) with:
@@ -57,6 +67,15 @@ They are executed on `unmetered locations` and as such won't consume your reques
 pytest tests
 ```
 
+## Docs
+
+From the directory run
+```bash
+mkdocs build
+mkdocs serve
+```
+In a browser navigate to `localhost:8000` to see the documentation.
+
 ### Formatters and Linters
 
 | Language | Formatter/Linter |

docs/api/pandafiableresponse.md

@@ -0,0 +1 @@
+::: solcast.api.PandafiableResponse

docs/index.md

@@ -1,21 +1,28 @@
 # Welcome to Solcast
-A simple Python SDK that wraps [Solcast's API](https://docs.solcast.com.au/). 
+A simple Python SDK that wraps [Solcast's API](https://docs.solcast.com.au/).
 
 ## Install
 From the directory run the following command:
-```bash 
+```bash
 pip install --user solcast
 ```
 !!! tip
 
-    for full functionality install **all**: `pip install --user solcast[all]`
+    for full functionality install **pandas**: `pip install --user solcast pandas`
+
+The example notebooks use a variety of optional dependencies to showcase different
+ways in which the Solcast API may be used. To install these dependencies run
+
+```commandline
+pip install --user solcast[all]
+```
 
 ## Usage
-!!! warning 
+!!! warning
 
     To access Solcast data you will need a [commercial API key](https://toolkit.solcast.com.au/register). If you have the API key already,
     you can use it with this library either as an environment variable called SOLCAST_API_KEY,
-    or you can pass it as an argument `api_key` when you call one of the library's methods. 
+    or you can pass it as an argument `api_key` when you call one of the library's methods.
 
 Fetching live radiation and weather data:
 
@@ -40,7 +47,7 @@ from solcast.unmetered_locations import UNMETERED_LOCATIONS
 sydney = UNMETERED_LOCATIONS['Sydney Opera House']
 
 res = forecast.rooftop_pv_power(
-    latitude=sydney['latitude'], 
+    latitude=sydney['latitude'],
     longitude=sydney['longitude'],
     period='PT5M',
     capacity=5,  # 5KW
@@ -49,22 +56,22 @@ res = forecast.rooftop_pv_power(
 )
 ```
 
-
-Where the data returned is a timeseries, the response can be converted to a Pandas DataFrame as follows.  This is available for all the modules apart from `pv_power_sites`.
+All response data can be extracted in Python dictionary format.
 
 ```python
-df = res.to_pandas()
+d = res.to_dict()
 ```
-!!! info
-    Pandas is not installed by default to keep the environment light. It is installed with the [all] tag
 
-For all the modules, data can be extracted in Python dictionary format.
+If pandas is installed, timeseries responses can be converted to a Pandas DataFrame using the ``.to_pandas()`` method.
+
 ```python
-df = res.to_dict()
+df = res.to_pandas()
 ```
+!!! info
+    Pandas is not installed by default to keep the environment light.
 
 
-Available modules are 
+Available modules are
 
 | Module           | API Docs                                 |
 |------------------|------------------------------------------|
@@ -76,14 +83,6 @@ Available modules are
 | `aggregations`   | [solcast.aggregations](aggregations.md)     |
 
 
-## Docs
-from the directory run
-```bash 
-mkdocs build
-mkdocs serve
-```
-In a browser navigate to `localhost:8000` to see the documentation.
-
 ## Contributing & License
 Any type of suggestion and code contribution is welcome as PRs and/or Issues.
 This repository is licensed under MIT (see LICENSE).

docs/notebooks/1.3 Getting Data - Make Concurrent Requests.ipynb

@@ -9,15 +9,6 @@
     "It is important to note that it is possible to exceed your rate limit as you increase the number of parallel downloads, which may cause some requests to fail!"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# ! pip install pandas matplotlib"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 1,
@@ -82,7 +73,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
    "outputs": [
     {
@@ -223,8 +214,8 @@
     "        df.append(res.to_pandas())\n",
     "    else:\n",
     "        # NOTE for production purposes you will need to deal with API failures, e.g. due rate-limiting!\n",
-    "        pass  \n",
-    "    \n",
+    "        pass\n",
+    "\n",
     "df = pd.concat(df)\n",
     "df"
    ]