粮食产量的估计¶

In [2]:

%load_ext autoreload
%autoreload 2
%matplotlib inline
%config InlineBackend.figure_format = 'retina'
from IPython.core.interactiveshell import InteractiveShell

InteractiveShell.ast_node_interactivity = "all"

from pathlib import Path
import pandas as pd
import numpy as np
from hydra import compose, initialize
import os

# 加载项目层面的配置
with initialize(version_base=None, config_path="../../config"):
    cfg = compose(config_name="config")
os.chdir(cfg.root)

%load_ext autoreload %autoreload 2 %matplotlib inline %config InlineBackend.figure_format = 'retina' from IPython.core.interactiveshell import InteractiveShell InteractiveShell.ast_node_interactivity = "all" from pathlib import Path import pandas as pd import numpy as np from hydra import compose, initialize import os # 加载项目层面的配置 with initialize(version_base=None, config_path="../../config"): cfg = compose(config_name="config") os.chdir(cfg.root)

In [3]:

from src.ci import ExpAnalyzer

multirun = ExpAnalyzer("out/2024-06-07/20-50-35")
multirun.overrides
multirun.diff_runs

from src.ci import ExpAnalyzer multirun = ExpAnalyzer("out/2024-06-07/20-50-35") multirun.overrides multirun.diff_runs

Out[3]:

{'farmer.forced_since': ['1987', '1998', '2020']}

Out[3]:

	farmer.forced_since
0	1998
1	2020
2	1987

In [4]:

from src.ci import ResultAnalyzer

datasets = []
for subdir in multirun.subdir:
    res = ResultAnalyzer(subdir)
    data = res.get_data(level="province", water_unit="m3")
    data["forced_since"] = res.select("farmer.forced_since")
    datasets.append(data)

df = pd.concat(datasets)

from src.ci import ResultAnalyzer datasets = [] for subdir in multirun.subdir: res = ResultAnalyzer(subdir) data = res.get_data(level="province", water_unit="m3") data["forced_since"] = res.select("farmer.forced_since") datasets.append(data) df = pd.concat(datasets)

In [15]:

data = df.groupby(["Time"])["capital_yield"].sum().reset_index()
data.head()

data = df.groupby(["Time"])["capital_yield"].sum().reset_index() data.head()

Out[15]:

	forced_since	Time	crop	capital_yield
0	1987	1981	Maize	1.340478e+09
1	1987	1981	Rice	3.843846e+06
2	1987	1981	Wheat	0.000000e+00
3	1987	1982	Maize	1.316527e+09
4	1987	1982	Rice	3.158475e+06

In [26]:

%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns

data = df.groupby(["Time", "crop"])["capital_yield"].sum().reset_index()
data.head()

sns.lineplot(
    data=data,
    x="Time",
    y="capital_yield",
    hue="crop",
    markers=True,
    dashes=True,
)

%matplotlib inline import matplotlib.pyplot as plt import seaborn as sns data = df.groupby(["Time", "crop"])["capital_yield"].sum().reset_index() data.head() sns.lineplot( data=data, x="Time", y="capital_yield", hue="crop", markers=True, dashes=True, )

Out[26]:

	Time	crop	capital_yield
0	1981	Maize	4.017307e+09
1	1981	Rice	1.149768e+07
2	1981	Wheat	0.000000e+00
3	1982	Maize	3.945145e+09
4	1982	Rice	9.429900e+06

Out[26]:

<Axes: xlabel='Time', ylabel='capital_yield'>

没有对粮食产量产生显著影响?

因为现在的灌溉策略，就是假设只要给的水够净灌溉量，都能够直接满足作物生长需求；