feat(random_map_generator): generate maps randomly

biomes/constants.py

@@ -1,12 +1,13 @@
 import os
+from pathlib import Path
 
-directory = os.path.dirname(os.path.abspath(__file__))
+directory = Path(os.path.dirname(os.path.abspath(__file__)))
 
-GEODATA = os.path.join(directory, 'geodata')
+GEODATA = directory / 'geodata'
-ECOREGIONS = os.path.join(GEODATA, 'ecoregions', 'single-parts.shp')
+ECOREGIONS = GEODATA / 'ecoregions' / 'single-parts.shp'
-ELEVATION = os.path.join(GEODATA, 'srtm', 'topo30-180.tif')
+ELEVATION = GEODATA / 'srtm' / 'topo30-180.tif'
-TEMP = os.path.join(GEODATA, 'air_temp')
+TEMP = GEODATA / 'air_temp'
-PRECIP = os.path.join(GEODATA, 'precipitation')
+PRECIP = GEODATA / 'precipitation'
 
 MIN_YEAR = 1900
 MAX_YEAR = 2017

biomes/draw.py

@@ -8,7 +8,7 @@ from constants import BIOMES
 import pandas as pd
 import cartopy.crs as ccrs
 
-def draw(df, earth=True, width=23.22, height=13, only_draw=False, path=None, alpha=1):
+def draw(df, earth=True, width=23.22, height=13, longitude_max=None, latitude_max=None, only_draw=False, path=None, alpha=1):
     logger.debug('draw(df, %s)', path)
     biomes = {}
     biome_numbers = df['biome_num'].unique()
@@ -17,7 +17,7 @@ def draw(df, earth=True, width=23.22, height=13, only_draw=False, path=None, alp
         if earth:
             p = (row.longitude, row.latitude)
         else:
-            p = (to_range(-180, 180, 0, width)(row.longitude), to_range(-90, 90, 0, height)(row.latitude))
+            p = (to_range(-180, 180, 0, longitude_max)(row.longitude), to_range(-90, 90, 0, latitude_max)(row.latitude))
 
         if row.biome_num in biomes:
             biomes[row.biome_num].append(p)

biomes/map_generator.py

@@ -57,7 +57,7 @@ parameters = {
         'max': 1e5,
     },
     'water_proportion': {
-        'default': 0.6,
+        'default': 0.8,
         'type': 'float',
         'min': 0,
         'max': 0.99,
@@ -242,7 +242,6 @@ def random_elevate_agent(ground, position, height, size=p['mountain_area_elevati
 
 
 def mountain_agent(ground, position):
-    print('mountain_agent')
     if not away_from_sea(ground, position):
         return
 
@@ -373,7 +372,7 @@ def generate_biomes(ground):
 
     print('running prediction models')
     print(p['mean_precipitation'], p['mean_temperature'])
-    result = predict_end_to_end(df, boundary)
+    result = predict_end_to_end(df)
 
     # fig = plt.figure()
     # ax = fig.add_subplot(111)
@@ -392,7 +391,7 @@ print(df['elevation'].min(), df['elevation'].max())
 print(df['distance_to_water'].min(), df['distance_to_water'].max())
 print(df['latitude'].min(), df['latitude'].max())
 
-def predict_end_to_end(input_df, boundary, checkpoint_temp='checkpoints/temp.h5', checkpoint_precip='checkpoints/precip.h5', checkpoint_biomes='checkpoints/b.h5', year=2000):
+def predict_end_to_end(input_df, checkpoint_temp='checkpoints/temp.h5', checkpoint_precip='checkpoints/precip.h5', checkpoint_biomes='checkpoints/b.h5', year=2000):
     batch_size = A_params['batch_size']['grid_search'][0]
     layers = A_params['layers']['grid_search'][0]
     optimizer = A_params['optimizer']['grid_search'][0](A_params['lr']['grid_search'][0])
@@ -428,8 +427,8 @@ def predict_end_to_end(input_df, boundary, checkpoint_temp='checkpoints/temp.h5'
     inputs = input_df[INPUTS]
 
     inputs.loc[:, 'mean_temp'] = p['mean_temperature']
-    inputs_copy = inputs.copy()
+    inputs_copy = df[INPUTS]
-    inputs_copy.loc[:, 'mean_temp'] = mean_temperature_over_years(df, size=inputs.shape[0])
+    inputs_copy.loc[:, 'mean_temp'] = mean_temperature_over_years(df, size=inputs_copy.shape[0])
 
     inputs = inputs.to_numpy()
     inputs = normalize_ndarray(inputs, inputs_copy)
@@ -440,8 +439,8 @@ def predict_end_to_end(input_df, boundary, checkpoint_temp='checkpoints/temp.h5'
     inputs = input_df[INPUTS]
 
     inputs.loc[:, 'mean_precip'] = p['mean_precipitation']
-    inputs_copy = inputs.copy()
+    inputs_copy = df[INPUTS]
-    inputs_copy.loc[:, 'mean_precip'] = mean_precipitation_over_years(df, size=inputs.shape[0])
+    inputs_copy.loc[:, 'mean_precip'] = mean_precipitation_over_years(df, size=inputs_copy.shape[0])
 
     inputs = inputs.to_numpy()
     inputs = normalize_ndarray(inputs, inputs_copy)
@@ -466,7 +465,7 @@ def predict_end_to_end(input_df, boundary, checkpoint_temp='checkpoints/temp.h5'
 
     columns = ['latitude', 'longitude', 'biome_num']
     new_data = pd.DataFrame(columns=columns)
-    nframe = pd.DataFrame(columns=frame.columns, data=normalize_ndarray(frame.to_numpy()))
+    nframe = pd.DataFrame(columns=frame.columns, data=normalize_ndarray(frame.to_numpy(), df[frame.columns].to_numpy()))
 
     for season in SEASONS:
         inputs += [
@@ -487,14 +486,10 @@ def predict_end_to_end(input_df, boundary, checkpoint_temp='checkpoints/temp.h5'
         }, columns=columns)
         new_data = new_data.append(f)
 
-    #print(new_data)
+    draw(new_data, earth=False, only_draw=True, longitude_max=p['width'], latitude_max=p['height'], alpha=0.7)
-    draw(new_data, earth=False, only_draw=True, width=p['width'], height=p['height'], alpha=0.7)
 
-# TODO: reduce opacity of biome layer
 if __name__ == "__main__":
     # p['width'] = 50
     # p['height'] = 50
-    p['seed'] = 3
     generate_map(True)
-    # print(normalize_ndarray(np.array([[ 5.59359803,0.99879546,-90., 45.24], [  5.54976747, 0.99879546,-86.4, 45.24      ]])))
     plt.show()

biomes/mars.py

@@ -0,0 +1,35 @@
+import rasterio
+from constants import *
+import matplotlib.pyplot as plt
+from map_generator import predict_end_to_end
+import pandas as pd
+
+MARS=GEODATA / 'planets' / 'Mars_MGS_MOLA_DEM_mosaic_global_463m.tif'
+elevation = rasterio.open(ELEVATION)
+elevation_data = elevation.read(1)
+
+
+data = {
+    'longitude': [],
+    'latitude': [],
+    'elevation': [],
+    'distance_to_water': []
+}
+
+for longitude in range(-179, 179):
+    print('-', end='')
+    for latitude in range(-89, 89):
+        data['longitude'].append(longitude)
+        data['latitude'].append(latitude)
+
+        elev = elevation_data[elevation.index(longitude, latitude)]
+        data['elevation'].append(elev)
+        data['distance_to_water'] = 999
+        print('+', end='')
+
+    print('')
+
+df = pd.DataFrame(data)
+
+predict_end_to_end(df)
+plt.show()

biomes/predict.py

@@ -138,15 +138,16 @@ def predicted_precips_cmd(checkpoint='checkpoints/precip.h5', year=2000):
     Precip.restore(checkpoint)
     predicted_precips(Precip, year=year)
 
-def predict_end_to_end(Temp, Precip, Biomes, year=2000):
+def predict_end_to_end(Temp, Precip, Biomes, df=pd.read_pickle('data.p'), year=2000):
     columns = INPUTS
 
-    df = pd.read_pickle('data.p')
-
     inputs = df[INPUTS]
+    earth_df = pd.read_pickle('data.p')
+    earth_df_inputs = earth_df[INPUTS]
 
     all_temps = ['temp_{}_{}'.format(season, year) for season in SEASONS]
     inputs.loc[:, 'mean_temp'] = np.mean(df[all_temps].values)
+    earth_df_inputs.loc[:, 'mean_temp'] = np.mean(earth_df[all_temps].values)
     print(inputs['mean_temp'])
 
     inputs = inputs.to_numpy()
@@ -208,7 +209,7 @@ def predict_end_to_end(Temp, Precip, Biomes, year=2000):
     print(new_data)
     draw(new_data)
 
-def predict_end_to_end_cmd(checkpoint_temp='checkpoints/temp.h5', checkpoint_precip='checkpoints/precip.h5', checkpoint_biomes='checkpoints/b.h5', year=2000):
+def predict_end_to_end_cmd(checkpoint_temp='checkpoints/temp.h5', checkpoint_precip='checkpoints/precip.h5', checkpoint_biomes='checkpoints/b.h5', year=2000, **kwargs):
     batch_size = A_params['batch_size']['grid_search'][0]
     layers = A_params['layers']['grid_search'][0]
     optimizer = A_params['optimizer']['grid_search'][0](A_params['lr']['grid_search'][0])
@@ -241,7 +242,7 @@ def predict_end_to_end_cmd(checkpoint_temp='checkpoints/temp.h5', checkpoint_pre
     Biomes.prepare_for_use()
     Biomes.restore(checkpoint_biomes)
 
-    predict_end_to_end(Temp=Temp, Precip=Precip, Biomes=Biomes, year=year)
+    predict_end_to_end(Temp=Temp, Precip=Precip, Biomes=Biomes, year=year, **kwargs)
 
 
 if __name__ == "__main__":

biomes/random_map_generator.py

@@ -0,0 +1,12 @@
+from map_generator import generate_map
+from pathlib import Path
+import matplotlib.pyplot as plt
+
+seed = 0
+while True:
+    generate_map(biomes=True, seed=seed)
+    root = Path('maps') / 'generated'
+    plt.savefig(root / f'{seed}.png')
+    plt.clf()
+
+    seed += 1