# Get available methods and properties for MadaclimLayers
>>> from py_madaclim.info import MadaclimLayers
>>> mada_info = MadaclimLayers()
>>> print(mada_info)
MadaclimLayers(
	all_layers = DataFrame(79 rows x 6 columns)
	categorical_layers = DataFrame(Layers 75, 76, 77, 78 with a total of 79 categories
	public methods -> download_data, fetch_specific_layers, get_categorical_combinations
			 get_layers_labels, select_geoclim_type_layers
)

# To access all layers as a dataframe
>>> mada_info.all_layers
geoclim_type  layer_number layer_name                       layer_description  is_categorical    units
0         clim             1      tmin1   Monthly minimum temperature - January           False  °C x 10
...

# Built-in method to download the Madaclim raster files
>>> mada_info.download_data(save_dir=cwd)

>>> env_labels = mada_info.get_layers_labels(
    layers_subset="env", 
    as_descriptive_labels=True
)
>>> print(env_labels[0])
'env_71_alt_Altitude (meters)'

>>> from py_madaclim.raster_manipulation import MadaclimRasters

>>> mada_rasters = MadaclimRasters("madaclim_current.tif", "madaclim_enviro.tif")
>>> print(mada_rasters)
MadaclimRasters(
	clim_raster = madaclim_current.tif,
	clim_crs = epsg:32738,
	clim_nodata_val = -32768.0
	env_raster = madaclim_enviro.tif,
	env_crs = epsg:32738,
	env_nodata_val = -32768.0
)

# Basic visualization for a continuous data layer
>>> mada_rasters.plot_layer(
...     layer=env_labels[0], 
...     imshow_cmap="terrain", 
...     histplot_binwidth=100, histplot_stat="count", 
... )

>>> from py_madaclim.raster_manipulation import MadaclimPoint

# Single point
>>> specimen_1 = MadaclimPoint(specimen_id="abbayesii", longitude=46.8624, latitude=-24.7541)

# Multipoints
>>> coll = MadaclimCollection.populate_from_csv("collection_example.csv")
>>> print(coll[0])
MadaclimPoint(
	specimen_id = ABA,
	source_crs = 4326,
	longitude = 46.8624,
	latitude = -24.7541,
	mada_geom_point = POINT (688328.2403248843 7260998.022932809),
	sampled_layers = None (Not sampled yet),
	nodata_layers = None (Not sampled yet),
	is_categorical_encoded = False,
	Species = C.abbayesii,
	Botanical_series = Millotii,
	Genome_size_2C_pg = 1.25,
	gdf.shape = (1, 11)
)

# Sample the collection reflects the changes to the geodataframe
>>> coll.sample_from_rasters(
...     clim_raster=mada_rasters.clim_raster,
...     env_raster=mada_rasters.env_raster,
...     layers_to_sample="all",   # Or any single/list of layers labels
...     layer_info=True
... )
>>> coll.gdf["specimen_id", env_labels[-1]]

# Visualize on the raster map
>>> coll.plot_on_layer(env_labels[-1], imshow_cmap="coolwarm")

# One Hot encoding updates the object dynamically
>>> coll.binary_encode_categorical()
>>> print(coll.is_categorical_encoded)
True
>>> coll_categ_layers = set(["_".join(label.split("_")[:4]) for label in coll.encoded_categ_labels])
>>> print(f"Splitted {len(coll_categ_layers)} layers into {len(coll.encoded_categ_labels)} unique categories")
Splitted 4 layers into 83 unique categories

# Updated geodataframe attribute
>>> env_76_encoded = coll.encoded_categ_labels[12:30]
>>> coll.gdf[["specimen_id"] + env_76_encoded]

>>> from py_madaclim.utils import gbif_api

# Get taxonKey of interest
>>> coffea_key = gbif_api.get_taxon_key_by_species_match("coffea")
EXACT match type found with 95% confidence!
canonical name of match: Coffea
GBIF_taxon_key: 2895315

# Search occurrences
recent_years = (2010, 2023)
>>> coffea_search_results_2010_present = gbif_api.search_occ_mdg_valid_coordinates(
...     taxon_key=coffea_key,
...     year_range=recent_years
... )
Fetching all 613 occurences in year range 2010-2023...
Extracting occurences 0 to 300...
Extracting occurences 300 to 600...
Extracting occurences 600 to 613...
Total records retrieved: 613

# ...Or create a download for a given search
>>> from dotenv import load_dotenv
>>> import os
>>> load_dotenv(".env")
True
>>> download_id = gbif_api.request_occ_download_mdg_valid_coordinates(
...     taxon_key=coffea_key,
...     email=your_email@gmail.com,
...     year_range=recent_years    # Defaults to None which is all possible years
... )

# Download, extract and read as df
>>> coffea_gbif_df = gbif_api.download_extract_read_occ(
    download_id=download_id,
    target_dir="gbif_example"
)
Response OK from https://api.gbif.org/v1/occurrence/download for the given 'download_id'
Progress for download_0008397-230810091245214.zip : 100.0% completed of 0.21 MB downloaded [ average speed of 0.41 MB/s ]
Extracting all 17 files to target location: .../download_0008397-230810091245214/
Read and saved core data into pandas df: occurrence.txt

# Keep relevant data
>>> df = coffea_gbif_df.loc[coffea_gbif_df["taxonRank"] == "SPECIES"]
>>> df = df.loc[:, ["verbatimScientificName", "decimalLongitude", "decimalLatitude", "year"]]
>>> df = df.reset_index().drop(columns="index")
>>> df["specimen_id"] = df.apply(lambda row: f"{row['verbatimScientificName']}_{row.name}", axis=1)
>>> df["specimen_id"] = df["specimen_id"].str.strip("Coffea ")
>>> # Format for MadaclimCollection constructor
>>> df.columns = ["genus_species", "longitude", "latitude", "year", "specimen_id"]
>>> df.head()

# Create a collection from recent samples
recent_coffea = MadaclimCollection.populate_from_df(df.loc[df["year"] >= 2020])