Data type and precision
python-float8,float16,float32,float64和float128可以包含多少位数？
np.finfo(np.float16).precision
# 3
np.finfo(np.float32).precision
# 6
np.finfo(np.float64).precision
# 15
np.finfo(np.float128).precision
Numpy slicing array (important!)
**Note: In numpy and NCL, rule of slicing array is: **
aa[i_start: i_end: slicing_step]
aa(i_start: i_end: slicing_step)
** But in MATLAB, it is different: **
aa(i_start: slicing_step: i_end)
** Do remeber it clearly or it will make mistakes!! **
Sort N-D numpy array based on one column
e.g. sorting a 2-cloumn array based on value of the first column by np.argsort
Z1=np.transpose(np.vstack((precp_Z1, d18Op_Z1)))
Z2=np.transpose(np.vstack((precp_Z2, d18Op_Z2)))
Z1_sort = HS1_Z1[np.argsort(Z1[:,0])]
Z2_sort = LGM_Z2[np.argsort(Z2[:,0])]
Initializing a numpy array of string data
The numpy string array is limited by its fixed length (default: length of 1). If the length of the strings are unsure or not equal, we can initialize the numpy string array by setting the data type as dtype=object:
newstr_array = numpy.empty((N, M), dtype=object)
Pythonic list comprehension
A short line of code to generate a list
[x * x for x in range(1, 11) if x % 2 == 0]
#which is identical to:
L = []
for x in range(1, 11):
if x % 2 ==0:
L.append(x * x)
More reference:
列表生成式
Read ascii files
Way 1: read()
Read() read all data at once and return a string.
with open("test.txt", "r") as f:
data = f.read()
print(data)
Way2: readline()
readline() read data by line and only return the first line. So loop would be used to read all lines of the data. Note that ‘\n’ is included!
with open("test.txt", "r") as f:
data = f.readline()
print(data)
Way3: readlines()
readlines() read data by lines and return all lines of data in a list. Note that ‘\n’ is included!
with open("test.txt", "r") as f:
data = f.readlines()
print(data)
Read data by lines and remove line break
Way1:
fnames = []
with open("need_files.txt") as fp:
#[fnames.append(i) for i in fp.readlines()]
[fnames.append(i) for i in fp.read().splitlines()]
Way2:
with open("need_files.txt", "r") as f:
data = f.readlines()
print(data)
need_data = []
need_data.append(i) for i in data[i][:-1]
Read and write data in numpy
Andes_loc = np.zeros((len(Andes_lat),2), dtype='int')
Andes_loc[:,0] = Andes_lon
Andes_loc[:,1] = Andes_lat
# output index
np.savetxt('Andes_locidx.txt', Andes_loc, fmt='%d')
# read data into an array
locIdx = np.loadtxt('Andes_locidx.txt', Andes_loc, fmt='%d')
Matplotlib colors
Choosing Colormaps in Matplotlib
Matplotlib special characters
Writing mathematical expressions
Matplotlib unaligned subplot (GridSpec) & double Y axis & bar plot
e.g.
import matplotlib.gridspec as gridspec
bar_width = 0.4
fig = plt.figure(figsize=(10,14))
gs = gridspec.GridSpec(4, 3) # subplot in 4*3 settings
ax1 = fig.add_subplot(gs[0,0])
c1 = ax1.bar(x=range(NM-1), height=Total_zone1[:-1], color='blue', alpha=0.6, width=bar_width)
c2 = ax1.bar(x=np.arange(NM-1)+bar_width, height=Total_zone1[:-1], color='red', alpha=0.6, width=bar_width)
ax12 = ax1.twinx()
c12 = ax12.bar(NM-1, height=Total_zone1[-1], color='blue', alpha=0.6, width=bar_width)
c22 = ax12.bar(NM-1+bar_width, height=Total_zone1[-1], color='red', alpha=0.6, width=bar_width)
Matplotlib: change blank size and subplot space
subplots_adjust(left=None, bottom=None, right=None, top=None,wspace=None,hspace=None) wspace：horizontal space between each subplot hspace：vertical space between each subplot
fig.tight_layout() #调整整体空白
plt.subplots_adjust(wspace =0, hspace =0)
Matplotlib broken Y axis
https://matplotlib.org/2.0.2/examples/pylab_examples/broken_axis.html
https://blog.csdn.net/Forrest97/article/details/113746307
seaborn CDF plot interacted with matplotlib
seaborn CDF subplots in a double y axis figrue
fig = plt.figure()
ax1 = fig.add_subplot(1,2,1)
ax1 = sns.distplot(precp_LGM_Z1,
kde_kws={'cumulative':'True', 'linestyle':'--', 'color':'blue'}, hist=False, ax=ax1)
ax1 = sns.distplot(precp_HS1_Z1,
kde_kws={'cumulative':'True', 'linestyle':'--', 'color':'red'}, hist=False, ax=ax1)
ax1.set_xlim([0, 35])
ax1.set_ylim([0, 1])
ax12 = ax1.twinx()
ax12.plot(LGM_Z1_draw[:,0], LGM_Z1_draw[:,1], color='blue')
ax12.plot(HS1_Z1_draw[:,0], HS1_Z1_draw[:,1], color='red')
ax12.set_ylim([-16, 0])
ax1.set_xlabel('mm/day')
ax1.set_ylabel('CDF')
ax1.set_title('(a) E Brazil', fontweight='bold')
ax2 = fig.add_subplot(1,2,2)
ax2 = sns.distplot(precp_LGM_Z2,
kde_kws={'cumulative':'True', 'linestyle':'--', 'color':'blue'}, hist=False, ax=ax2)
ax2 = sns.distplot(precp_HS1_Z2,
kde_kws={'cumulative':'True', 'linestyle':'--', 'color':'red'}, hist=False, ax=ax2)
ax2.set_xlim([0, 35])
ax2.set_ylim([0, 1])
ax2.set_ylabel('')
ax2.set_xlabel('mm/day')
ax22 = ax2.twinx()
c1 = ax22.plot(LGM_Z2_draw[:,0], LGM_Z2_draw[:,1], color='blue')
ax22.plot(HS1_Z2_draw[:,0], HS1_Z2_draw[:,1], color='red')
ax22.set_ylim([-16, 0])
ax22.set_ylabel('$\delta^{18}O_p$ (permil)')
ax2.set_title('(b) W Amazon', fontweight='bold')