notebook.community

Edit and run 





In [3]:



    


%run ./msgc_experiments_CMBBEIV_init.ipynb



    


















    






Populating the interactive namespace from numpy and matplotlib

















In [4]:



    


import pandas as pd
import seaborn as sns

Data processing graphs, statistics

Seaborn graphs





In [5]:



    


df = pd.read_csv(fnamenew)
df["image voxel number"] = df["data size"].pow(3)
df["relative object size"] = df["data object size px"] / df["image voxel number"]
df["log(relative object size)"] = log(df["data object size px"] / df["image voxel number"])
# df.rename(columns={"msgc time": "MSGC time"})
# dfs = df[(df["data seedsz"]==3) & (df["data offset"] == 3) & (df["data radius"] == 10) & (df["experiment"] == "image size")]
dfs = df[df["experiment"] == "image size"]
dfs_plus = dfs[dfs['data size'] > 160]

sns.set_context("paper")
sns.set_style("white")



    











In [6]:



    


df.keys()



    


















    
Out[6]:








Index(['_create_nlinks time', 'block size', 'data image size px',
       'data object size px', 'data offset', 'data radius', 'data seedsz',
       'data size', 'edge number', 'error', 'experiment',
       'experiment iteration start time', 'gc time', 'low level image voxels',
       'low level object voxels', 'machine hostname', 'method',
       'nlinks number', 't graph 01', 't graph 10', 't graph 11', 't graph 13',
       't graph 14', 't graph high', 't graph low', 't split 01', 't split 02',
       't split 03', 't split 04', 't split 05', 't split 06', 't split 07',
       't split 08', 't split 081', 't split 082', 't split 0821',
       't split 09', 't split 10', 't1', 't10', 't2', 't3', 't3.1', 't3.2',
       't3.3', 't4', 't5', 't6', 't7', 't8', 't9', 'time', 'tlinks number',
       'image voxel number', 'relative object size',
       'log(relative object size)'],
      dtype='object')

















In [7]:



    


sns.boxplot(data=dfs_plus, y="time", x="method", showfliers=False)



    


















    
Out[7]:








<matplotlib.axes._subplots.AxesSubplot at 0x1e720351dd8>










    
























In [8]:



    


# df
mfmc_label = "Min-Cut"
rename_bp = {"gc time": mfmc_label, "time": "total"}



    











In [9]:



    


uu = pd.melt(dfs_plus.rename(columns=rename_bp), value_vars=[mfmc_label, "total"], id_vars=["method"], var_name="time type", value_name="Time [s]")
# uu = pd.melt(dfs, value_vars=["gc time", "time"], id_vars=["method"], var_name="type", value_name="time")
# uu

Paper





In [10]:



    


mfmc_label = "Min-Cut"
rename_bp = {"gc time": mfmc_label, "time": "total"}

bp = sns.boxplot(data=uu, hue="time type",y="Time [s]", x="method", 
#                  showfliers=False
                 hue_order=["total", mfmc_label],
                )
# bp.
bp.set(yscale="log")
# ax.set(xscale="log", yscale="log")
plt.savefig(op.join(latex_dir, "msgc_time_boxplot.pdf"), dpi=1000)



    


















    
























In [11]:



    


bp.set(yscale="log")



    


















    
Out[11]:








[None]

















In [12]:



    


# sns.boxplot(data=dfs, y="error", x="method")



    











In [13]:



    


lm = sns.lmplot(data=df, x="data size", y="time", hue="method", order=3, scatter_kws={"s": 3, "marker": "x", "alpha": 0.5})
axes = lm.axes
axes[0,0].set_xlim(30,200)
axes[0,0].set_ylim(0,50)

lines = lm.ax.get_lines()
line = lines[0]
line.set_linestyle("--")
# line.set_marker("s")

plt.savefig(op.join(latex_dir, "msgc_time_datasize_plot.pdf"), dpi=1000)
# axes[0,1].set_ylim(0,)

# lm.ax.get_lines()

Vykreslení dvou lmplotů do jednoho není tak snadné





In [14]:



    


# test better melt

dfm = pd.DataFrame({'City': ['Houston', 'Austin', 'Hoover'],
                   'State': ['Texas', 'Texas', 'Alabama'],
                   'Name':['Aria', 'Penelope', 'Niko'],
                   'Mango':[4, 10, 90],
                   'Orange': [10, 8, 14], 
                   'Watermelon':[40, 99, 43],
                   'Gin':[16, 200, 34],
                   'Vodka':[20, 33, 18]},
                 columns=['City', 'State', 'Name', 'Mango', 'Orange', 'Watermelon', 'Gin', 'Vodka'])
melt(dfm, id_vars=['City', 'State'], value_vars=[['Mango', 'Orange', 'Watermelon'], ['Gin', 'Vodka']], 
     var_name=['Fruit', 'Drink'], value_name=['Pounds', 'Ounces'])



    


















    
Out[14]:











  
    

      
      City
      State
      Fruit
      Pounds
      Drink
      Ounces
    

  
  
    

      0
      Houston
      Texas
      Mango
      4
      Gin
      16.0
    

    

      1
      Austin
      Texas
      Mango
      10
      Gin
      200.0
    

    

      2
      Hoover
      Alabama
      Mango
      90
      Gin
      34.0
    

    

      3
      Houston
      Texas
      Orange
      10
      Vodka
      20.0
    

    

      4
      Austin
      Texas
      Orange
      8
      Vodka
      33.0
    

    

      5
      Hoover
      Alabama
      Orange
      14
      Vodka
      18.0
    

    

      6
      Houston
      Texas
      Watermelon
      40
      nan
      NaN
    

    

      7
      Austin
      Texas
      Watermelon
      99
      nan
      NaN
    

    

      8
      Hoover
      Alabama
      Watermelon
      43
      nan
      NaN
    

  




















In [15]:



    


# df.keys()



    











In [16]:



    


# df.rename(columns={"gc time": "gc", "time": "total"})[["gc", "total", "data size"]]



    











In [17]:



    


# df.rename(columns={"gc time": "gc", "time": "total"})



    











In [18]:



    


dfs = df[df["experiment"] == "image size"]
uu = melt(dfs.rename(columns={"gc time": "gc", "time": "total"}), value_vars=["gc", "total"], id_vars=["method", "data size", "image voxel number", "relative object size"], var_name=["time type"], value_name=["Time [s]"])
uu["mth"] = uu["method"] + " " + uu["time type"]
# uu



    











In [19]:



    


current_palette = sns.color_palette()
sns.palplot(current_palette)

Artifical data - constant object size





In [20]:



    


white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(data=uu, x="data size", y="Time [s]", hue="mth", order=3, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3]
                palette=white_palette
               )
axes = lm.axes
axes[0,0].set_xlim(30,200)
axes[0,0].set_ylim(0,50)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())



    


















    
























In [21]:



    


white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu, 
    x="data size", y="Time [s]", hue="mth", order=3, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3],
                palette=white_palette,
                legend=None,
#     legend="off"
               )
axes = lm.axes
axes[0,0].set_xlim(30,200)
axes[0,0].set_ylim(0,50)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
sc = sns.scatterplot(data=uu[uu["time type"] == "total"], x="data size", y="Time [s]", hue="mth", ax=lm.ax, alpha=0.8, s=3)
texts = lm.ax.legend_.get_texts()
leglines = lm.ax.legend_.get_lines()

# texts[0].set_text("short")
# texts.pop(0)
# texts.pop(0)

# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
lm.ax.legend(
    custom_lines, 
    [tx.get_text()[:-3] for tx in texts[:3]],
#     ['Cold', 'Medium', 'Hot'],
)
# lg = lm.ax.legend()



    


















    
Out[21]:








<matplotlib.legend.Legend at 0x1e7219cc2b0>










    
























In [22]:



    


texts[0].get_text()



    


















    
Out[22]:








'ssgc  gc'

















In [23]:



    


# dir(lm.ax.legend_)



    











In [24]:



    


# L=plt.legend()
lm.ax.legend_.get_texts()
# L.get_texts()[0].set_text('make it short')



    


















    
Out[24]:








<a list of 3 Text objects>

Paper - With voxel number





In [25]:



    


rename_time = {
    "time": "Time [s]"
}
# uu = 
current_palette = sns.color_palette()
white_palette = [
    (1., 1., 1., 0.5),
    (1., 1., 1., 0.5),
    (1., 1., 1., 0.5),
    current_palette[0],
    current_palette[1],
    current_palette[2],
]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu, 
    x="image voxel number", y="Time [s]", hue="mth", order=3, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3],
                palette=white_palette,
                legend=None,
#     legend="off"
               )
axes = lm.axes
axes[0,0].set_xlim(30,uu["image voxel number"].max())
axes[0,0].set_ylim(0,50)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
# sc = sns.scatterplot(
#     data=uu[uu["time type"] == "total"], x="image voxel number", y="time", hue="mth", 
#     ax=lm.ax, alpha=1.0, s=5
# )
# texts = lm.ax.legend_.get_texts()
# leglines = lm.ax.legend_.get_lines()


# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
# lm.ax.legend(
#     custom_lines, 
#     [tx.get_text()[:-3] for tx in texts[:3]],
# )
# lg = lm.ax.legend()
lm.ax.legend(
    custom_lines, 
    labels
#     [tx.get_text()[:-3] for tx in texts[:3]],
)


plt.savefig(op.join(latex_dir, "msgc_time_image_voxel_number_plot.pdf"), dpi=1000)



    


















    
























In [26]:



    


uu



    


















    
Out[26]:











  
    

      
      method
      data size
      image voxel number
      relative object size
      time type
      Time [s]
      mth
    

  
  
    

      0
      ssgc
      44
      85184
      0.093727
      gc
      0.330146
      ssgc  gc
    

    

      1
      msgc_hi2lo
      44
      85184
      0.093727
      gc
      0.371008
      msgc_hi2lo  gc
    

    

      2
      msgc_lo2hi
      44
      85184
      0.093727
      gc
      0.548532
      msgc_lo2hi  gc
    

    

      3
      ssgc
      44
      85184
      0.093727
      gc
      0.327159
      ssgc  gc
    

    

      4
      msgc_hi2lo
      44
      85184
      0.093727
      gc
      0.382976
      msgc_hi2lo  gc
    

    

      5
      msgc_lo2hi
      44
      85184
      0.093727
      gc
      0.466750
      msgc_lo2hi  gc
    

    

      6
      ssgc
      47
      103823
      0.078615
      gc
      0.387964
      ssgc  gc
    

    

      7
      msgc_hi2lo
      47
      103823
      0.078615
      gc
      0.539556
      msgc_hi2lo  gc
    

    

      8
      msgc_lo2hi
      47
      103823
      0.078615
      gc
      0.534571
      msgc_lo2hi  gc
    

    

      9
      ssgc
      47
      103823
      0.078615
      gc
      0.459755
      ssgc  gc
    

    

      10
      msgc_hi2lo
      47
      103823
      0.078615
      gc
      0.540556
      msgc_hi2lo  gc
    

    

      11
      msgc_lo2hi
      47
      103823
      0.078615
      gc
      0.477721
      msgc_lo2hi  gc
    

    

      12
      ssgc
      50
      125000
      0.068992
      gc
      0.444810
      ssgc  gc
    

    

      13
      msgc_hi2lo
      50
      125000
      0.068992
      gc
      0.562531
      msgc_hi2lo  gc
    

    

      14
      msgc_lo2hi
      50
      125000
      0.068992
      gc
      0.606413
      msgc_lo2hi  gc
    

    

      15
      ssgc
      50
      125000
      0.068992
      gc
      0.444811
      ssgc  gc
    

    

      16
      msgc_hi2lo
      50
      125000
      0.068992
      gc
      0.670211
      msgc_hi2lo  gc
    

    

      17
      msgc_lo2hi
      50
      125000
      0.068992
      gc
      0.674193
      msgc_lo2hi  gc
    

    

      18
      ssgc
      53
      148877
      0.060728
      gc
      0.620343
      ssgc  gc
    

    

      19
      msgc_hi2lo
      53
      148877
      0.060728
      gc
      0.822799
      msgc_hi2lo  gc
    

    

      20
      msgc_lo2hi
      53
      148877
      0.060728
      gc
      0.726058
      msgc_lo2hi  gc
    

    

      21
      ssgc
      53
      148877
      0.058988
      gc
      0.499663
      ssgc  gc
    

    

      22
      msgc_hi2lo
      53
      148877
      0.058988
      gc
      0.772935
      msgc_hi2lo  gc
    

    

      23
      msgc_lo2hi
      53
      148877
      0.058988
      gc
      0.701123
      msgc_lo2hi  gc
    

    

      24
      ssgc
      56
      175616
      0.051129
      gc
      0.563528
      ssgc  gc
    

    

      25
      msgc_hi2lo
      56
      175616
      0.051129
      gc
      0.702153
      msgc_hi2lo  gc
    

    

      26
      msgc_lo2hi
      56
      175616
      0.051129
      gc
      0.754981
      msgc_lo2hi  gc
    

    

      27
      ssgc
      56
      175616
      0.051129
      gc
      0.641288
      ssgc  gc
    

    

      28
      msgc_hi2lo
      56
      175616
      0.051129
      gc
      0.854715
      msgc_hi2lo  gc
    

    

      29
      msgc_lo2hi
      56
      175616
      0.051129
      gc
      0.979398
      msgc_lo2hi  gc
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      918
      ssgc
      152
      3511808
      0.005038
      total
      13.689396
      ssgc  total
    

    

      919
      msgc_hi2lo
      152
      3511808
      0.005038
      total
      15.491609
      msgc_hi2lo  total
    

    

      920
      msgc_lo2hi
      152
      3511808
      0.005038
      total
      10.887919
      msgc_lo2hi  total
    

    

      921
      ssgc
      155
      3723875
      0.004850
      total
      15.438683
      ssgc  total
    

    

      922
      msgc_hi2lo
      155
      3723875
      0.004850
      total
      16.157799
      msgc_hi2lo  total
    

    

      923
      msgc_lo2hi
      155
      3723875
      0.004850
      total
      8.189136
      msgc_lo2hi  total
    

    

      924
      ssgc
      158
      3944312
      0.004744
      total
      16.157797
      ssgc  total
    

    

      925
      msgc_hi2lo
      158
      3944312
      0.004744
      total
      17.363541
      msgc_hi2lo  total
    

    

      926
      msgc_lo2hi
      158
      3944312
      0.004744
      total
      12.214330
      msgc_lo2hi  total
    

    

      927
      ssgc
      161
      4173281
      0.004376
      total
      17.411446
      ssgc  total
    

    

      928
      msgc_hi2lo
      161
      4173281
      0.004376
      total
      18.864563
      msgc_hi2lo  total
    

    

      929
      msgc_lo2hi
      161
      4173281
      0.004376
      total
      10.693441
      msgc_lo2hi  total
    

    

      930
      ssgc
      164
      4410944
      0.004366
      total
      18.455324
      ssgc  total
    

    

      931
      msgc_hi2lo
      164
      4410944
      0.004366
      total
      19.835914
      msgc_hi2lo  total
    

    

      932
      msgc_lo2hi
      164
      4410944
      0.004366
      total
      13.916789
      msgc_lo2hi  total
    

    

      933
      ssgc
      167
      4657463
      0.004270
      total
      21.011818
      ssgc  total
    

    

      934
      msgc_hi2lo
      167
      4657463
      0.004270
      total
      20.820364
      msgc_hi2lo  total
    

    

      935
      msgc_lo2hi
      167
      4657463
      0.004270
      total
      15.304080
      msgc_lo2hi  total
    

    

      936
      ssgc
      170
      4913000
      0.003961
      total
      22.141764
      ssgc  total
    

    

      937
      msgc_hi2lo
      170
      4913000
      0.003961
      total
      21.030734
      msgc_hi2lo  total
    

    

      938
      msgc_lo2hi
      170
      4913000
      0.003961
      total
      14.966982
      msgc_lo2hi  total
    

    

      939
      ssgc
      173
      5177717
      0.003841
      total
      24.428684
      ssgc  total
    

    

      940
      msgc_hi2lo
      173
      5177717
      0.003841
      total
      22.747177
      msgc_hi2lo  total
    

    

      941
      msgc_lo2hi
      173
      5177717
      0.003841
      total
      15.659095
      msgc_lo2hi  total
    

    

      942
      ssgc
      176
      5451776
      0.003760
      total
      26.212910
      ssgc  total
    

    

      943
      msgc_hi2lo
      176
      5451776
      0.003760
      total
      23.712629
      msgc_hi2lo  total
    

    

      944
      msgc_lo2hi
      176
      5451776
      0.003760
      total
      16.071030
      msgc_lo2hi  total
    

    

      945
      ssgc
      179
      5735339
      0.003694
      total
      30.277045
      ssgc  total
    

    

      946
      msgc_hi2lo
      179
      5735339
      0.003694
      total
      24.703947
      msgc_hi2lo  total
    

    

      947
      msgc_lo2hi
      179
      5735339
      0.003694
      total
      17.161115
      msgc_lo2hi  total
    

  



948 rows × 7 columns



















In [27]:



    


xx = "relative object size"
white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu, 
    x=xx, y="Time [s]", hue="mth", 
#     order=2, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3],
                palette=white_palette,
                legend=None,
                logx=True
#     legend="off"
               )
axes = lm.axes
# axes[0,0].set_xlim(30,uu[xx].max())
axes[0,0].set_ylim(0,50)
# lm.set(
#     xscale="log",
# #     yscale="log"
# )

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
sc = sns.scatterplot(
    data=uu[uu["time type"] == "total"], x=xx, y="Time [s]", hue="mth", 
    ax=lm.ax, alpha=1.0, s=5
)
texts = lm.ax.legend_.get_texts()
leglines = lm.ax.legend_.get_lines()

# texts[0].set_text("short")
# texts.pop(0)
# texts.pop(0)

# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
lm.ax.legend(
    custom_lines, 
    [tx.get_text()[:-3] for tx in texts[:3]],
#     ['Cold', 'Medium', 'Hot'],
)
# lg = lm.ax.legend()



    


















    






C:\Users\Jirik\Miniconda3\envs\lisa\lib\site-packages\seaborn\regression.py:279: RuntimeWarning: invalid value encountered in log
  grid = np.c_[np.ones(len(grid)), np.log(grid)]










    
Out[27]:








<matplotlib.legend.Legend at 0x1e721a401d0>










    
























In [28]:



    


sc = sns.scatterplot(
    data=uu[uu["time type"] == "total"], x=xx, y="Time [s]", hue="mth", 
#     ax=lm.ax,
    alpha=1.0, s=5
)
sc.set(xscale="log",
      yscale="log"
      )
# grid.set(xscale="log", )



    


















    
Out[28]:








[None, None]










    
























In [29]:



    


lm.ax



    


















    
Out[29]:








<matplotlib.axes._subplots.AxesSubplot at 0x1e721925908>

















In [30]:



    


dir(lm.ax)



    


















    
Out[30]:








['__class__',
 '__delattr__',
 '__dict__',
 '__dir__',
 '__doc__',
 '__eq__',
 '__format__',
 '__ge__',
 '__getattribute__',
 '__getstate__',
 '__gt__',
 '__hash__',
 '__init__',
 '__init_subclass__',
 '__le__',
 '__lt__',
 '__module__',
 '__ne__',
 '__new__',
 '__reduce__',
 '__reduce_ex__',
 '__repr__',
 '__setattr__',
 '__setstate__',
 '__sizeof__',
 '__str__',
 '__subclasshook__',
 '__weakref__',
 '_add_text',
 '_adjustable',
 '_agg_filter',
 '_alpha',
 '_anchor',
 '_animated',
 '_aspect',
 '_autoscaleXon',
 '_autoscaleYon',
 '_autotitlepos',
 '_axes',
 '_axes_class',
 '_axes_locator',
 '_axisbelow',
 '_cachedRenderer',
 '_clipon',
 '_clippath',
 '_connected',
 '_contains',
 '_current_image',
 '_facecolor',
 '_frameon',
 '_gci',
 '_gen_axes_patch',
 '_gen_axes_spines',
 '_get_axis_list',
 '_get_lines',
 '_get_patches_for_fill',
 '_get_view',
 '_gid',
 '_gridOn',
 '_hold',
 '_in_layout',
 '_init_axis',
 '_label',
 '_layoutbox',
 '_left_title',
 '_make_twin_axes',
 '_mouseover',
 '_mouseover_set',
 '_navigate',
 '_navigate_mode',
 '_oid',
 '_on_units_changed',
 '_originalPosition',
 '_path_effects',
 '_pcolorargs',
 '_picker',
 '_position',
 '_poslayoutbox',
 '_process_unit_info',
 '_prop_order',
 '_propobservers',
 '_quiver_units',
 '_rasterization_zorder',
 '_rasterized',
 '_remove_legend',
 '_remove_method',
 '_right_title',
 '_sci',
 '_set_artist_props',
 '_set_gc_clip',
 '_set_lim_and_transforms',
 '_set_position',
 '_set_title_offset_trans',
 '_set_view',
 '_set_view_from_bbox',
 '_shared_x_axes',
 '_shared_y_axes',
 '_sharex',
 '_sharey',
 '_sketch',
 '_snap',
 '_stale',
 '_sticky_edges',
 '_subplotspec',
 '_tight',
 '_transform',
 '_transformSet',
 '_twinned_axes',
 '_update_image_limits',
 '_update_line_limits',
 '_update_patch_limits',
 '_update_title_position',
 '_update_transScale',
 '_url',
 '_use_sticky_edges',
 '_validate_converted_limits',
 '_visible',
 '_xaxis_transform',
 '_xcid',
 '_xmargin',
 '_yaxis_transform',
 '_ycid',
 '_ymargin',
 'acorr',
 'add_artist',
 'add_callback',
 'add_child_axes',
 'add_collection',
 'add_container',
 'add_image',
 'add_line',
 'add_patch',
 'add_table',
 'aname',
 'angle_spectrum',
 'annotate',
 'apply_aspect',
 'arrow',
 'artists',
 'autoscale',
 'autoscale_view',
 'axes',
 'axhline',
 'axhspan',
 'axis',
 'axison',
 'axvline',
 'axvspan',
 'bar',
 'barbs',
 'barh',
 'bbox',
 'boxplot',
 'broken_barh',
 'bxp',
 'callbacks',
 'can_pan',
 'can_zoom',
 'change_geometry',
 'child_axes',
 'cla',
 'clabel',
 'clear',
 'clipbox',
 'cohere',
 'colNum',
 'collections',
 'containers',
 'contains',
 'contains_point',
 'contour',
 'contourf',
 'convert_xunits',
 'convert_yunits',
 'csd',
 'dataLim',
 'drag_pan',
 'draw',
 'draw_artist',
 'end_pan',
 'errorbar',
 'eventplot',
 'eventson',
 'figbox',
 'figure',
 'fill',
 'fill_between',
 'fill_betweenx',
 'findobj',
 'fmt_xdata',
 'fmt_ydata',
 'format_coord',
 'format_cursor_data',
 'format_xdata',
 'format_ydata',
 'get_adjustable',
 'get_agg_filter',
 'get_alpha',
 'get_anchor',
 'get_animated',
 'get_aspect',
 'get_autoscale_on',
 'get_autoscalex_on',
 'get_autoscaley_on',
 'get_axes_locator',
 'get_axisbelow',
 'get_children',
 'get_clip_box',
 'get_clip_on',
 'get_clip_path',
 'get_contains',
 'get_cursor_data',
 'get_data_ratio',
 'get_data_ratio_log',
 'get_default_bbox_extra_artists',
 'get_facecolor',
 'get_fc',
 'get_figure',
 'get_frame_on',
 'get_geometry',
 'get_gid',
 'get_gridspec',
 'get_images',
 'get_in_layout',
 'get_label',
 'get_legend',
 'get_legend_handles_labels',
 'get_lines',
 'get_navigate',
 'get_navigate_mode',
 'get_path_effects',
 'get_picker',
 'get_position',
 'get_rasterization_zorder',
 'get_rasterized',
 'get_renderer_cache',
 'get_shared_x_axes',
 'get_shared_y_axes',
 'get_sketch_params',
 'get_snap',
 'get_subplotspec',
 'get_tightbbox',
 'get_title',
 'get_transform',
 'get_transformed_clip_path_and_affine',
 'get_url',
 'get_visible',
 'get_window_extent',
 'get_xaxis',
 'get_xaxis_text1_transform',
 'get_xaxis_text2_transform',
 'get_xaxis_transform',
 'get_xbound',
 'get_xgridlines',
 'get_xlabel',
 'get_xlim',
 'get_xmajorticklabels',
 'get_xminorticklabels',
 'get_xscale',
 'get_xticklabels',
 'get_xticklines',
 'get_xticks',
 'get_yaxis',
 'get_yaxis_text1_transform',
 'get_yaxis_text2_transform',
 'get_yaxis_transform',
 'get_ybound',
 'get_ygridlines',
 'get_ylabel',
 'get_ylim',
 'get_ymajorticklabels',
 'get_yminorticklabels',
 'get_yscale',
 'get_yticklabels',
 'get_yticklines',
 'get_yticks',
 'get_zorder',
 'grid',
 'has_data',
 'have_units',
 'hexbin',
 'hist',
 'hist2d',
 'hitlist',
 'hlines',
 'ignore_existing_data_limits',
 'images',
 'imshow',
 'in_axes',
 'indicate_inset',
 'indicate_inset_zoom',
 'inset_axes',
 'invert_xaxis',
 'invert_yaxis',
 'is_figure_set',
 'is_first_col',
 'is_first_row',
 'is_last_col',
 'is_last_row',
 'is_transform_set',
 'label_outer',
 'legend',
 'legend_',
 'lines',
 'locator_params',
 'loglog',
 'magnitude_spectrum',
 'margins',
 'matshow',
 'minorticks_off',
 'minorticks_on',
 'mouseover',
 'mouseover_set',
 'name',
 'numCols',
 'numRows',
 'patch',
 'patches',
 'pchanged',
 'pcolor',
 'pcolorfast',
 'pcolormesh',
 'phase_spectrum',
 'pick',
 'pickable',
 'pie',
 'plot',
 'plot_date',
 'properties',
 'psd',
 'quiver',
 'quiverkey',
 'redraw_in_frame',
 'relim',
 'remove',
 'remove_callback',
 'reset_position',
 'rowNum',
 'scatter',
 'semilogx',
 'semilogy',
 'set',
 'set_adjustable',
 'set_agg_filter',
 'set_alpha',
 'set_anchor',
 'set_animated',
 'set_aspect',
 'set_autoscale_on',
 'set_autoscalex_on',
 'set_autoscaley_on',
 'set_axes_locator',
 'set_axis_off',
 'set_axis_on',
 'set_axisbelow',
 'set_clip_box',
 'set_clip_on',
 'set_clip_path',
 'set_contains',
 'set_facecolor',
 'set_fc',
 'set_figure',
 'set_frame_on',
 'set_gid',
 'set_in_layout',
 'set_label',
 'set_navigate',
 'set_navigate_mode',
 'set_path_effects',
 'set_picker',
 'set_position',
 'set_prop_cycle',
 'set_rasterization_zorder',
 'set_rasterized',
 'set_sketch_params',
 'set_snap',
 'set_subplotspec',
 'set_title',
 'set_transform',
 'set_url',
 'set_visible',
 'set_xbound',
 'set_xlabel',
 'set_xlim',
 'set_xmargin',
 'set_xscale',
 'set_xticklabels',
 'set_xticks',
 'set_ybound',
 'set_ylabel',
 'set_ylim',
 'set_ymargin',
 'set_yscale',
 'set_yticklabels',
 'set_yticks',
 'set_zorder',
 'specgram',
 'spines',
 'spy',
 'stackplot',
 'stale',
 'stale_callback',
 'start_pan',
 'stem',
 'step',
 'sticky_edges',
 'streamplot',
 'table',
 'tables',
 'text',
 'texts',
 'tick_params',
 'ticklabel_format',
 'title',
 'titleOffsetTrans',
 'transAxes',
 'transData',
 'transLimits',
 'transScale',
 'tricontour',
 'tricontourf',
 'tripcolor',
 'triplot',
 'twinx',
 'twiny',
 'update',
 'update_datalim',
 'update_datalim_bounds',
 'update_from',
 'update_params',
 'use_sticky_edges',
 'viewLim',
 'violin',
 'violinplot',
 'vlines',
 'xaxis',
 'xaxis_date',
 'xaxis_inverted',
 'xcorr',
 'yaxis',
 'yaxis_date',
 'yaxis_inverted',
 'zorder']

















In [31]:



    


lm = sns.lmplot(data=uu, x="data size", y="Time [s]", hue="time type", order=3, scatter_kws={"s": 3, "marker": "x", "alpha": 0.5})
axes = lm.axes
axes[0,0].set_xlim(30,200)
axes[0,0].set_ylim(0,50)

lines = lm.ax.get_lines()
line = lines[0]
line.set_linestyle("--")
# line.set_marker("s")

# plt.savefig(op.join(latex_dir, "msgc_time_datasize_plot.pdf"), dpi=1000)
# axes[0,1].set_ylim(0,)

# lm.ax.get_lines()



    


















    
























In [ ]:



    






    











In [32]:



    


# this work jus for not duplicit values of data siz
# sns.tsplot(data=df, time="data size", value="time", unit="method", condition="method")
# plt.savefig(op.join(latex_dir, "msgc_size_time.pdf"), dpi=1000)



    











In [33]:



    


line.set_marker("s")



    











In [34]:



    


# df



    











In [35]:



    


# df["method"]

Experiment 2: fixed image size





In [36]:



    


# uu = df[df["experiment"]=="object size"]



    











In [37]:



    


dfs2 = df[df["experiment"] == "object size"]
uu2 = melt(dfs2.rename(columns={"gc time": "gc", "time": "total"}), value_vars=["gc", "total"], id_vars=["method", "data size", "image voxel number", "data object size px", "relative object size", "log(relative object size)"], var_name=["time type"], value_name=["Time [s]"])
uu2["mth"] = uu2["method"] + " " + uu2["time type"]
# uu



    











In [38]:



    


dfs2.keys()



    


















    
Out[38]:








Index(['_create_nlinks time', 'block size', 'data image size px',
       'data object size px', 'data offset', 'data radius', 'data seedsz',
       'data size', 'edge number', 'error', 'experiment',
       'experiment iteration start time', 'gc time', 'low level image voxels',
       'low level object voxels', 'machine hostname', 'method',
       'nlinks number', 't graph 01', 't graph 10', 't graph 11', 't graph 13',
       't graph 14', 't graph high', 't graph low', 't split 01', 't split 02',
       't split 03', 't split 04', 't split 05', 't split 06', 't split 07',
       't split 08', 't split 081', 't split 082', 't split 0821',
       't split 09', 't split 10', 't1', 't10', 't2', 't3', 't3.1', 't3.2',
       't3.3', 't4', 't5', 't6', 't7', 't8', 't9', 'time', 'tlinks number',
       'image voxel number', 'relative object size',
       'log(relative object size)'],
      dtype='object')

















In [39]:



    


xx="data object size px"
current_palette = sns.color_palette()
white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu2, 
#     x="image voxel number", 
    x=xx,
    y="Time [s]", hue="mth", order=2, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
                palette=current_palette[:3],
#                 palette=white_palette,
                legend=None,
#     legend="off"
               )
axes = lm.axes
axes[0,0].set_xlim(uu2[xx].min(), uu2[xx].max())
axes[0,0].set_ylim(0, 400)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
sc = sns.scatterplot(
    data=uu2[uu2["time type"] == "total"], x=xx, y="Time [s]", hue="mth", 
    ax=lm.ax, alpha=1.0, s=5
)
texts = lm.ax.legend_.get_texts()
leglines = lm.ax.legend_.get_lines()

# texts[0].set_text("short")
# texts.pop(0)
# texts.pop(0)

# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
lm.ax.legend(
    custom_lines, 
    [tx.get_text()[:-3] for tx in texts[:3]],
#     ['Cold', 'Medium', 'Hot'],
)
# lg = lm.ax.legend()
plt.savefig(op.join(latex_dir, "msgc_time_object_size_plot.pdf"), dpi=1000)



    


















    
























In [40]:



    


uu2



    


















    
Out[40]:











  
    

      
      method
      data size
      image voxel number
      data object size px
      relative object size
      log(relative object size)
      time type
      Time [s]
      mth
    

  
  
    

      0
      ssgc
      200
      8000000
      5654
      0.000707
      -7.254834
      gc
      31.700240
      ssgc  gc
    

    

      1
      msgc_hi2lo
      200
      8000000
      5654
      0.000707
      -7.254834
      gc
      1.687488
      msgc_hi2lo  gc
    

    

      2
      msgc_lo2hi
      200
      8000000
      5654
      0.000707
      -7.254834
      gc
      1.157905
      msgc_lo2hi  gc
    

    

      3
      ssgc
      200
      8000000
      14722
      0.001840
      -6.297854
      gc
      31.278368
      ssgc  gc
    

    

      4
      msgc_hi2lo
      200
      8000000
      14722
      0.001840
      -6.297854
      gc
      2.115344
      msgc_hi2lo  gc
    

    

      5
      msgc_lo2hi
      200
      8000000
      14722
      0.001840
      -6.297854
      gc
      1.322432
      msgc_lo2hi  gc
    

    

      6
      ssgc
      200
      8000000
      27598
      0.003450
      -5.669454
      gc
      33.441616
      ssgc  gc
    

    

      7
      msgc_hi2lo
      200
      8000000
      27598
      0.003450
      -5.669454
      gc
      2.467438
      msgc_hi2lo  gc
    

    

      8
      msgc_lo2hi
      200
      8000000
      27598
      0.003450
      -5.669454
      gc
      1.549857
      msgc_lo2hi  gc
    

    

      9
      ssgc
      200
      8000000
      47289
      0.005911
      -5.130919
      gc
      33.118446
      ssgc  gc
    

    

      10
      msgc_hi2lo
      200
      8000000
      47289
      0.005911
      -5.130919
      gc
      3.179531
      msgc_hi2lo  gc
    

    

      11
      msgc_lo2hi
      200
      8000000
      47289
      0.005911
      -5.130919
      gc
      1.990675
      msgc_lo2hi  gc
    

    

      12
      ssgc
      200
      8000000
      72397
      0.009050
      -4.705032
      gc
      32.783346
      ssgc  gc
    

    

      13
      msgc_hi2lo
      200
      8000000
      72397
      0.009050
      -4.705032
      gc
      13.494916
      msgc_hi2lo  gc
    

    

      14
      msgc_lo2hi
      200
      8000000
      72397
      0.009050
      -4.705032
      gc
      2.986015
      msgc_lo2hi  gc
    

    

      15
      ssgc
      200
      8000000
      107141
      0.013393
      -4.313051
      gc
      33.363792
      ssgc  gc
    

    

      16
      msgc_hi2lo
      200
      8000000
      107141
      0.013393
      -4.313051
      gc
      4.495979
      msgc_hi2lo  gc
    

    

      17
      msgc_lo2hi
      200
      8000000
      107141
      0.013393
      -4.313051
      gc
      2.995989
      msgc_lo2hi  gc
    

    

      18
      ssgc
      200
      8000000
      150317
      0.018790
      -3.974450
      gc
      34.987450
      ssgc  gc
    

    

      19
      msgc_hi2lo
      200
      8000000
      150317
      0.018790
      -3.974450
      gc
      5.845373
      msgc_hi2lo  gc
    

    

      20
      msgc_lo2hi
      200
      8000000
      150317
      0.018790
      -3.974450
      gc
      3.303167
      msgc_lo2hi  gc
    

    

      21
      ssgc
      200
      8000000
      202377
      0.025297
      -3.677065
      gc
      41.066231
      ssgc  gc
    

    

      22
      msgc_hi2lo
      200
      8000000
      202377
      0.025297
      -3.677065
      gc
      6.505608
      msgc_hi2lo  gc
    

    

      23
      msgc_lo2hi
      200
      8000000
      202377
      0.025297
      -3.677065
      gc
      3.911508
      msgc_lo2hi  gc
    

    

      24
      ssgc
      200
      8000000
      266438
      0.033305
      -3.402055
      gc
      35.604800
      ssgc  gc
    

    

      25
      msgc_hi2lo
      200
      8000000
      266438
      0.033305
      -3.402055
      gc
      7.966698
      msgc_hi2lo  gc
    

    

      26
      msgc_lo2hi
      200
      8000000
      266438
      0.033305
      -3.402055
      gc
      4.460077
      msgc_lo2hi  gc
    

    

      27
      ssgc
      200
      8000000
      339024
      0.042378
      -3.161126
      gc
      33.340855
      ssgc  gc
    

    

      28
      msgc_hi2lo
      200
      8000000
      339024
      0.042378
      -3.161126
      gc
      8.867294
      msgc_hi2lo  gc
    

    

      29
      msgc_lo2hi
      200
      8000000
      339024
      0.042378
      -3.161126
      gc
      5.694742
      msgc_lo2hi  gc
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      270
      ssgc
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      48.354741
      ssgc  total
    

    

      271
      msgc_hi2lo
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      31.445954
      msgc_hi2lo  total
    

    

      272
      msgc_lo2hi
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      12.425824
      msgc_lo2hi  total
    

    

      273
      ssgc
      200
      8000000
      56107
      0.007013
      -4.959936
      total
      43.702777
      ssgc  total
    

    

      274
      msgc_hi2lo
      200
      8000000
      56107
      0.007013
      -4.959936
      total
      34.284363
      msgc_hi2lo  total
    

    

      275
      msgc_lo2hi
      200
      8000000
      56107
      0.007013
      -4.959936
      total
      28.776059
      msgc_lo2hi  total
    

    

      276
      ssgc
      200
      8000000
      182047
      0.022756
      -3.782932
      total
      42.243080
      ssgc  total
    

    

      277
      msgc_hi2lo
      200
      8000000
      182047
      0.022756
      -3.782932
      total
      39.491440
      msgc_hi2lo  total
    

    

      278
      msgc_lo2hi
      200
      8000000
      182047
      0.022756
      -3.782932
      total
      79.791653
      msgc_lo2hi  total
    

    

      279
      ssgc
      200
      8000000
      428533
      0.053567
      -2.926829
      total
      43.246367
      ssgc  total
    

    

      280
      msgc_hi2lo
      200
      8000000
      428533
      0.053567
      -2.926829
      total
      43.210464
      msgc_hi2lo  total
    

    

      281
      msgc_lo2hi
      200
      8000000
      428533
      0.053567
      -2.926829
      total
      171.862476
      msgc_lo2hi  total
    

    

      282
      ssgc
      200
      8000000
      816188
      0.102024
      -2.282552
      total
      44.889974
      ssgc  total
    

    

      283
      msgc_hi2lo
      200
      8000000
      816188
      0.102024
      -2.282552
      total
      54.154266
      msgc_hi2lo  total
    

    

      284
      msgc_lo2hi
      200
      8000000
      816188
      0.102024
      -2.282552
      total
      349.147755
      msgc_lo2hi  total
    

    

      285
      ssgc
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      43.001019
      ssgc  total
    

    

      286
      msgc_hi2lo
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      31.801001
      msgc_hi2lo  total
    

    

      287
      msgc_lo2hi
      200
      8000000
      5654
      0.000707
      -7.254834
      total
      9.312102
      msgc_lo2hi  total
    

    

      288
      ssgc
      200
      8000000
      55085
      0.006886
      -4.978319
      total
      40.379034
      ssgc  total
    

    

      289
      msgc_hi2lo
      200
      8000000
      55085
      0.006886
      -4.978319
      total
      34.881733
      msgc_hi2lo  total
    

    

      290
      msgc_lo2hi
      200
      8000000
      55085
      0.006886
      -4.978319
      total
      28.357177
      msgc_lo2hi  total
    

    

      291
      ssgc
      200
      8000000
      179201
      0.022400
      -3.798689
      total
      41.308549
      ssgc  total
    

    

      292
      msgc_hi2lo
      200
      8000000
      179201
      0.022400
      -3.798689
      total
      37.511697
      msgc_hi2lo  total
    

    

      293
      msgc_lo2hi
      200
      8000000
      179201
      0.022400
      -3.798689
      total
      65.575669
      msgc_lo2hi  total
    

    

      294
      ssgc
      200
      8000000
      423029
      0.052879
      -2.939756
      total
      42.425529
      ssgc  total
    

    

      295
      msgc_hi2lo
      200
      8000000
      423029
      0.052879
      -2.939756
      total
      47.291965
      msgc_hi2lo  total
    

    

      296
      msgc_lo2hi
      200
      8000000
      423029
      0.052879
      -2.939756
      total
      160.700321
      msgc_lo2hi  total
    

    

      297
      ssgc
      200
      8000000
      806947
      0.100868
      -2.293939
      total
      44.661588
      ssgc  total
    

    

      298
      msgc_hi2lo
      200
      8000000
      806947
      0.100868
      -2.293939
      total
      51.946110
      msgc_hi2lo  total
    

    

      299
      msgc_lo2hi
      200
      8000000
      806947
      0.100868
      -2.293939
      total
      353.250604
      msgc_lo2hi  total
    

  



300 rows × 9 columns



















In [41]:



    


xx = "relative object size"
white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu2, 
#     x="image voxel number", 
    x=xx,
    y="Time [s]", hue="mth", 
#     order=1, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3],
                palette=white_palette,
                legend=None,
    logx=True,
#     logy=True
#     legend="off"
               )
axes = lm.axes
# axes[0,0].set_xlim(30,uu2[xx].max())
# axes[0,0].set_ylim(0,50)
# yscale="log"

lm.set(
    xscale="log",
# #     yscale="log"
)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
sc = sns.scatterplot(
    data=uu2[uu2["time type"] == "total"], x=xx, y="Time [s]", hue="mth", 
    ax=lm.ax, alpha=1.0, s=5
)
texts = lm.ax.legend_.get_texts()
leglines = lm.ax.legend_.get_lines()

# texts[0].set_text("short")
# texts.pop(0)
# texts.pop(0)

# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
lm.ax.legend(
    custom_lines, 
    [tx.get_text()[:-3] for tx in texts[:3]],
#     ['Cold', 'Medium', 'Hot'],
)
# lg = lm.ax.legend()
# plt.savefig(op.join(latex_dir, "msgc_time_datasize_plot.pdf"), dpi=1000)



    


















    
Out[41]:








<matplotlib.legend.Legend at 0x1e721dcce48>










    
























In [42]:



    


xx = "log(relative object size)"
white_palette = [(1., 1., 1., 0.5)]
lm = sns.lmplot(
#     data=uu[uu["time type"] == "total"], 
    data=uu2, 
#     x="image voxel number", 
    x=xx,
    y="Time [s]", hue="mth", order=2, 
                scatter_kws={"s": 3, "marker": "x", "alpha": 0.5}, 
#                 line_kws={"alpha": 1},
#                 palette=current_palette[:3],
                palette=white_palette,
                legend=None,
#     legend="off"
               )
axes = lm.axes
axes[0,0].set_xlim(uu2[xx].min(),uu2[xx].max())
axes[0,0].set_ylim(0,50)

current_palette = sns.color_palette()
nlines = 3
n = 0
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 1
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
# line1.set_color(line2.get_color())

n = 2
lines = lm.ax.get_lines()
line1 = lines[n]
line2 = lines[n + nlines]
line1.set_linestyle("--")
line1.set_color(current_palette[n])
line2.set_color(current_palette[n])
line1.set_color(line2.get_color())


# body
sc = sns.scatterplot(
    data=uu2[uu2["time type"] == "total"], x=xx, y="Time [s]", hue="mth", 
    ax=lm.ax, alpha=1.0, s=5
)
texts = lm.ax.legend_.get_texts()
leglines = lm.ax.legend_.get_lines()

# texts[0].set_text("short")
# texts.pop(0)
# texts.pop(0)

# plt.legend(lm.ax.legend_[3:6])

from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=current_palette[0], lw=4),
                Line2D([0], [0], color=current_palette[1], lw=4),
                Line2D([0], [0], color=current_palette[2], lw=4)]
lm.ax.legend(
    custom_lines, 
    [tx.get_text()[:-3] for tx in texts[:3]],
#     ['Cold', 'Medium', 'Hot'],
)
# lg = lm.ax.legend()
# plt.savefig(op.join(latex_dir, "msgc_time_datasize_plot.pdf"), dpi=1000)



    


















    
Out[42]:








<matplotlib.legend.Legend at 0x1e721e6a438>










    
























In [43]:



    


## Volume ration

Statistics





In [44]:



    


from scipy import stats



dfs_plus_describe = dfs_plus.describe()
display(dfs_plus_describe)

print("pokud je pvalue mensi nez zvolena hladina vyznamnosti (0.01=1%), je vsechno ok")
tt = stats.ttest_rel(dfs_plus.loc[dfs_plus["method"] == "ssgc"]['time'], dfs_plus.loc[dfs_plus["method"] == "msgc_lo2hi"]['time'])
# tt



    


















    











  
    

      
      _create_nlinks time
      block size
      data image size px
      data object size px
      data offset
      data radius
      data seedsz
      data size
      edge number
      error
      ...
      t5
      t6
      t7
      t8
      t9
      time
      tlinks number
      image voxel number
      relative object size
      log(relative object size)
    

  
  
    

      count
      123.000000
      123.0
      1.230000e+02
      123.000000
      123.000000
      123.0
      123.0
      123.000000
      1.230000e+02
      123.0
      ...
      82.000000
      82.000000
      82.000000
      41.000000
      82.000000
      123.000000
      1.230000e+02
      1.230000e+02
      123.000000
      123.000000
    

    

      mean
      1.460365
      10.0
      6.292836e+06
      20970.365854
      0.040000
      10.0
      3.0
      182.439024
      1.006576e+07
      0.0
      ...
      7.812621
      10.439942
      11.246214
      22.396002
      17.912613
      23.084985
      2.521442e+06
      6.191968e+06
      0.003519
      -5.662245
    

    

      std
      1.356393
      0.0
      1.534372e+06
      1790.517286
      0.009144
      0.0
      0.0
      14.795406
      1.122510e+07
      0.0
      ...
      5.555318
      3.722275
      3.146358
      5.885085
      6.381470
      9.096514
      2.820764e+06
      1.517977e+06
      0.000560
      0.161019
    

    

      min
      0.001994
      10.0
      4.251366e+06
      18263.000000
      0.030000
      10.0
      3.0
      161.000000
      2.023006e+06
      0.0
      ...
      1.636625
      4.761268
      6.134597
      14.872235
      9.103665
      10.693441
      5.013340e+05
      4.173281e+06
      0.002604
      -5.950865
    

    

      25%
      0.003990
      10.0
      5.000040e+06
      19462.000000
      0.030000
      10.0
      3.0
      170.000000
      2.474007e+06
      0.0
      ...
      2.511262
      7.271526
      8.897183
      17.652762
      13.006748
      16.092971
      6.133455e+05
      4.913000e+06
      0.003025
      -5.800855
    

    

      50%
      1.253683
      10.0
      5.831820e+06
      20699.000000
      0.040000
      10.0
      3.0
      179.000000
      2.712393e+06
      0.0
      ...
      6.784342
      10.393687
      11.245465
      20.948951
      16.427074
      20.940974
      6.725010e+05
      5.735339e+06
      0.003503
      -5.654180
    

    

      75%
      2.356696
      10.0
      7.414680e+06
      22437.000000
      0.050000
      10.0
      3.0
      194.000000
      1.991244e+07
      0.0
      ...
      12.635707
      12.989767
      13.256064
      25.679296
      20.968905
      26.544525
      5.000040e+06
      7.301384e+06
      0.003961
      -5.531176
    

    

      max
      5.320740
      10.0
      9.260790e+06
      24638.000000
      0.050000
      10.0
      3.0
      209.000000
      3.691086e+07
      0.0
      ...
      19.852917
      20.329675
      20.329675
      36.605654
      36.647542
      62.877877
      9.260790e+06
      9.129329e+06
      0.004473
      -5.409593
    

  



8 rows × 53 columns












    






pokud je pvalue mensi nez zvolena hladina vyznamnosti (0.01=1%), je vsechno ok

















In [45]:



    


ssgc_rows = dfs_plus[dfs_plus["method"].str.contains(labels[0])]
ssgc_hi2lo_rows = dfs_plus[dfs_plus["method"].str.contains(labels[1])]
ssgc_lo2hi_rows = dfs_plus[dfs_plus["method"].str.contains(labels[2])]
pp0 = stats.ttest_rel(ssgc_rows["time"], ssgc_hi2lo_rows["time"])
pp1 = stats.ttest_rel(ssgc_rows["time"], ssgc_lo2hi_rows["time"])
pp2 = stats.ttest_rel(ssgc_hi2lo_rows["time"], ssgc_lo2hi_rows["time"])

print("pokud je pvalue mensi nez zvolena hladina vyznamnosti (0.01=1%), je vsechno ok") 
#mozna staci i dvojnasobek hladiny vyzamnosi
print("statistic musi byt vetsi nez 0")
display(pp0)
display(pp1)
display(pp2)

float_to_latex_file(pp0.pvalue, op.join(latex_dir, "ttest_pvalue_ssgc_msgc_hi2lo.tex"))
float_to_latex_file(pp1.pvalue, op.join(latex_dir, "ttest_pvalue_ssgc_msgc_lo2hi.tex"))
float_to_latex_file(pp2.pvalue, op.join(latex_dir, "ttest_pvalue_msgc_hi2lo_msgc_lo2hi.tex"))



    


















    






pokud je pvalue mensi nez zvolena hladina vyznamnosti (0.01=1%), je vsechno ok
statistic musi byt vetsi nez 0










    








Ttest_relResult(statistic=4.197809231906577, pvalue=0.00014586011815140358)










    








Ttest_relResult(statistic=9.241651101451831, pvalue=1.784740931504529e-11)










    








Ttest_relResult(statistic=13.712192453675819, pvalue=1.0458893651364037e-16)

















In [46]:



    


# dfs_plus["method"] == "ssgc "

Time evaluation





In [47]:



    


table = pd.pivot_table(
    dfs, values=['gc time', 'time', "t2", "t3", "t3.1", "t3.2", "t3.3", "t4", "t5", "t6", "t7", "t8", "low level object voxels", "low level image voxels"], index=['method'], 
    aggfunc=np.mean
)
table



    


















    
Out[47]:











  
    

      
      gc time
      low level image voxels
      low level object voxels
      t2
      t3
      t3.1
      t3.2
      t3.3
      t4
      t5
      t6
      t7
      t8
      time
    

    

      method
      
      
      
      
      
      
      
      
      
      
      
      
      
      
    

  
  
    

      msgc_hi2lo
      1.805255
      2916.303797
      13.525316
      0.066423
      0.068504
      NaN
      NaN
      NaN
      0.524395
      1.112787
      3.228815
      4.565952
      10.152585
      12.210139
    

    

      msgc_lo2hi
      1.244942
      2916.303797
      11.601266
      0.082775
      0.084915
      0.084915
      0.106336
      7.377492
      7.377492
      8.203736
      8.340086
      8.340092
      NaN
      9.866378
    

    

      ssgc
      8.874517
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      11.190694
    

  




















In [48]:



    


table = pd.pivot_table(
    dfs, values=['gc time', 'time', "t graph 01", "t graph 10", "t graph 11", "t graph 13", "t graph 14", "low level image voxels", "low level object voxels"], index=['method'], 
    aggfunc=np.mean
)
table



    


















    
Out[48]:











  
    

      
      gc time
      low level image voxels
      low level object voxels
      t graph 01
      t graph 10
      t graph 11
      t graph 13
      t graph 14
      time
    

    

      method
      
      
      
      
      
      
      
      
      
    

  
  
    

      msgc_hi2lo
      1.805255
      2916.303797
      13.525316
      NaN
      NaN
      NaN
      NaN
      NaN
      12.210139
    

    

      msgc_lo2hi
      1.244942
      2916.303797
      11.601266
      0.020815
      0.021415
      0.021497
      7.213204
      7.292571
      9.866378
    

    

      ssgc
      8.874517
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      11.190694

Datasets





In [49]:



    


dfs_describe = dfs.describe()
display(dfs_describe)
dfs_plus_describe = dfs_plus.describe()
display(dfs_plus_describe)



    


















    











  
    

      
      _create_nlinks time
      block size
      data image size px
      data object size px
      data offset
      data radius
      data seedsz
      data size
      edge number
      error
      ...
      t5
      t6
      t7
      t8
      t9
      time
      tlinks number
      image voxel number
      relative object size
      log(relative object size)
    

  
  
    

      count
      474.000000
      474.0
      4.740000e+02
      474.000000
      474.000000
      474.0
      474.0
      474.000000
      4.740000e+02
      474.0
      ...
      316.000000
      316.000000
      316.000000
      158.000000
      316.000000
      474.000000
      4.740000e+02
      4.740000e+02
      474.000000
      474.000000
    

    

      mean
      0.628041
      10.0
      2.677025e+06
      14931.170886
      0.040000
      10.0
      3.0
      122.132911
      4.853243e+06
      0.0
      ...
      4.658261
      5.784450
      6.453022
      10.152585
      9.264756
      11.089070
      1.215826e+06
      2.625507e+06
      0.017022
      -4.599966
    

    

      std
      0.926779
      0.0
      2.491407e+06
      4517.543448
      0.008428
      0.0
      0.0
      46.685775
      7.073115e+06
      0.0
      ...
      4.614842
      4.422958
      4.168487
      8.524849
      6.765083
      9.210574
      1.776536e+06
      2.456507e+06
      0.020462
      0.975121
    

    

      min
      0.000953
      10.0
      9.108000e+04
      7984.000000
      0.030000
      10.0
      3.0
      44.000000
      3.582460e+05
      0.0
      ...
      0.032879
      0.085778
      0.335139
      0.475728
      0.475730
      0.332147
      9.108000e+04
      8.518400e+04
      0.002604
      -5.950865
    

    

      25%
      0.003011
      10.0
      5.926200e+05
      11207.000000
      0.030000
      10.0
      3.0
      83.000000
      1.666590e+06
      0.0
      ...
      0.684423
      1.696689
      2.785044
      3.691631
      4.201267
      4.400461
      4.155102e+05
      5.717870e+05
      0.004473
      -5.409593
    

    

      50%
      0.215392
      10.0
      1.860744e+06
      14455.500000
      0.040000
      10.0
      3.0
      122.000000
      2.281190e+06
      0.0
      ...
      2.540690
      5.323267
      5.858352
      7.313445
      8.370618
      9.157033
      5.653340e+05
      1.815848e+06
      0.008163
      -4.808105
    

    

      75%
      0.857464
      10.0
      4.251366e+06
      18669.000000
      0.050000
      10.0
      3.0
      161.000000
      2.887670e+06
      0.0
      ...
      8.456605
      9.070142
      9.351213
      14.938057
      12.258962
      14.932854
      7.170060e+05
      4.173281e+06
      0.019675
      -3.928398
    

    

      max
      5.320740
      10.0
      9.260790e+06
      24638.000000
      0.050000
      10.0
      3.0
      209.000000
      3.691086e+07
      0.0
      ...
      19.852917
      20.329675
      20.329675
      36.605654
      36.647542
      62.877877
      9.260790e+06
      9.129329e+06
      0.095816
      -2.345324
    

  



8 rows × 53 columns












    











  
    

      
      _create_nlinks time
      block size
      data image size px
      data object size px
      data offset
      data radius
      data seedsz
      data size
      edge number
      error
      ...
      t5
      t6
      t7
      t8
      t9
      time
      tlinks number
      image voxel number
      relative object size
      log(relative object size)
    

  
  
    

      count
      123.000000
      123.0
      1.230000e+02
      123.000000
      123.000000
      123.0
      123.0
      123.000000
      1.230000e+02
      123.0
      ...
      82.000000
      82.000000
      82.000000
      41.000000
      82.000000
      123.000000
      1.230000e+02
      1.230000e+02
      123.000000
      123.000000
    

    

      mean
      1.460365
      10.0
      6.292836e+06
      20970.365854
      0.040000
      10.0
      3.0
      182.439024
      1.006576e+07
      0.0
      ...
      7.812621
      10.439942
      11.246214
      22.396002
      17.912613
      23.084985
      2.521442e+06
      6.191968e+06
      0.003519
      -5.662245
    

    

      std
      1.356393
      0.0
      1.534372e+06
      1790.517286
      0.009144
      0.0
      0.0
      14.795406
      1.122510e+07
      0.0
      ...
      5.555318
      3.722275
      3.146358
      5.885085
      6.381470
      9.096514
      2.820764e+06
      1.517977e+06
      0.000560
      0.161019
    

    

      min
      0.001994
      10.0
      4.251366e+06
      18263.000000
      0.030000
      10.0
      3.0
      161.000000
      2.023006e+06
      0.0
      ...
      1.636625
      4.761268
      6.134597
      14.872235
      9.103665
      10.693441
      5.013340e+05
      4.173281e+06
      0.002604
      -5.950865
    

    

      25%
      0.003990
      10.0
      5.000040e+06
      19462.000000
      0.030000
      10.0
      3.0
      170.000000
      2.474007e+06
      0.0
      ...
      2.511262
      7.271526
      8.897183
      17.652762
      13.006748
      16.092971
      6.133455e+05
      4.913000e+06
      0.003025
      -5.800855
    

    

      50%
      1.253683
      10.0
      5.831820e+06
      20699.000000
      0.040000
      10.0
      3.0
      179.000000
      2.712393e+06
      0.0
      ...
      6.784342
      10.393687
      11.245465
      20.948951
      16.427074
      20.940974
      6.725010e+05
      5.735339e+06
      0.003503
      -5.654180
    

    

      75%
      2.356696
      10.0
      7.414680e+06
      22437.000000
      0.050000
      10.0
      3.0
      194.000000
      1.991244e+07
      0.0
      ...
      12.635707
      12.989767
      13.256064
      25.679296
      20.968905
      26.544525
      5.000040e+06
      7.301384e+06
      0.003961
      -5.531176
    

    

      max
      5.320740
      10.0
      9.260790e+06
      24638.000000
      0.050000
      10.0
      3.0
      209.000000
      3.691086e+07
      0.0
      ...
      19.852917
      20.329675
      20.329675
      36.605654
      36.647542
      62.877877
      9.260790e+06
      9.129329e+06
      0.004473
      -5.409593
    

  



8 rows × 53 columns

Dataset size





In [50]:



    


dfs_plus_size = int(len(dfs_plus) / len(labels))
to_file(str(dfs_plus_size), op.join(latex_dir, "msgc_dataset_subset_size.tex"))



    











In [51]:



    


dfs2_size = len(dfs2[dfs2["method"] == labels[0]])
to_file(str(dfs_plus_size), op.join(latex_dir, "msgc_dataset_subset2_size.tex"))



    











In [52]:



    


# df[["GC total time", "MSGC total time", "GC time", "MSGC time"]]
df.keys()



    


















    
Out[52]:








Index(['_create_nlinks time', 'block size', 'data image size px',
       'data object size px', 'data offset', 'data radius', 'data seedsz',
       'data size', 'edge number', 'error', 'experiment',
       'experiment iteration start time', 'gc time', 'low level image voxels',
       'low level object voxels', 'machine hostname', 'method',
       'nlinks number', 't graph 01', 't graph 10', 't graph 11', 't graph 13',
       't graph 14', 't graph high', 't graph low', 't split 01', 't split 02',
       't split 03', 't split 04', 't split 05', 't split 06', 't split 07',
       't split 08', 't split 081', 't split 082', 't split 0821',
       't split 09', 't split 10', 't1', 't10', 't2', 't3', 't3.1', 't3.2',
       't3.3', 't4', 't5', 't6', 't7', 't8', 't9', 'time', 'tlinks number',
       'image voxel number', 'relative object size',
       'log(relative object size)'],
      dtype='object')

















In [57]:



    


table = pd.pivot_table(
    df, values=['gc time', 'time', "error"],
    index=['method'], 
    aggfunc={'gc time': np.mean, 'time': [min, max, np.mean], "error":[min, max, np.mean]}
)
table



    


















    
Out[57]:











  
    

      
      error
      gc time
      time
    

    

      
      max
      mean
      min
      mean
      max
      mean
      min
    

    

      method
      
      
      
      
      
      
      
    

  
  
    

      msgc_hi2lo
      2.0
      0.038462
      0.0
      3.324061
      70.414725
      19.704389
      0.790861
    

    

      msgc_lo2hi
      2.0
      0.038462
      0.0
      2.242018
      412.448155
      38.454763
      0.825794
    

    

      ssgc
      2.0
      0.038462
      0.0
      15.911705
      76.873458
      19.459985
      0.332147
    

  




















In [54]:



    


dfs.keys()



    


















    
Out[54]:








Index(['_create_nlinks time', 'block size', 'data image size px',
       'data object size px', 'data offset', 'data radius', 'data seedsz',
       'data size', 'edge number', 'error', 'experiment',
       'experiment iteration start time', 'gc time', 'low level image voxels',
       'low level object voxels', 'machine hostname', 'method',
       'nlinks number', 't graph 01', 't graph 10', 't graph 11', 't graph 13',
       't graph 14', 't graph high', 't graph low', 't split 01', 't split 02',
       't split 03', 't split 04', 't split 05', 't split 06', 't split 07',
       't split 08', 't split 081', 't split 082', 't split 0821',
       't split 09', 't split 10', 't1', 't10', 't2', 't3', 't3.1', 't3.2',
       't3.3', 't4', 't5', 't6', 't7', 't8', 't9', 'time', 'tlinks number',
       'image voxel number', 'relative object size',
       'log(relative object size)'],
      dtype='object')

















In [55]:



    


df_mn = df[["GC total time", "MSGC total time", "GC time", "MSGC time"]].describe()
display(df_mn)
to_latex_file(df_mn, "../includes/exp062-all2data_size.tex")

dfs_mn = dfs[["GC total time", "MSGC total time", "GC time", "MSGC time"]].describe()
display(dfs_mn)
to_latex_file(dfs_mn, "../includes/exp062-selection2data_size.tex")



    


















    






---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-55-e7e71b79ce86> in <module>
----> 1 df_mn = df[["GC total time", "MSGC total time", "GC time", "MSGC time"]].describe()
      2 display(df_mn)
      3 to_latex_file(df_mn, "../includes/exp062-all2data_size.tex")
      4 
      5 dfs_mn = dfs[["GC total time", "MSGC total time", "GC time", "MSGC time"]].describe()

~\Miniconda3\envs\lisa\lib\site-packages\pandas\core\frame.py in __getitem__(self, key)
   2932                 key = list(key)
   2933             indexer = self.loc._convert_to_indexer(key, axis=1,
-> 2934                                                    raise_missing=True)
   2935 
   2936         # take() does not accept boolean indexers

~\Miniconda3\envs\lisa\lib\site-packages\pandas\core\indexing.py in _convert_to_indexer(self, obj, axis, is_setter, raise_missing)
   1352                 kwargs = {'raise_missing': True if is_setter else
   1353                           raise_missing}
-> 1354                 return self._get_listlike_indexer(obj, axis, **kwargs)[1]
   1355         else:
   1356             try:

~\Miniconda3\envs\lisa\lib\site-packages\pandas\core\indexing.py in _get_listlike_indexer(self, key, axis, raise_missing)
   1159         self._validate_read_indexer(keyarr, indexer,
   1160                                     o._get_axis_number(axis),
-> 1161                                     raise_missing=raise_missing)
   1162         return keyarr, indexer
   1163 

~\Miniconda3\envs\lisa\lib\site-packages\pandas\core\indexing.py in _validate_read_indexer(self, key, indexer, axis, raise_missing)
   1244                 raise KeyError(
   1245                     u"None of [{key}] are in the [{axis}]".format(
-> 1246                         key=key, axis=self.obj._get_axis_name(axis)))
   1247 
   1248             # We (temporarily) allow for some missing keys with .loc, except in

KeyError: "None of [Index(['GC total time', 'MSGC total time', 'GC time', 'MSGC time'], dtype='object')] are in the [columns]"
















In [ ]:



    


dfs_plus[["method"]]

Content source: mjirik/pyseg_base

Similar notebooks:

notebook.community | gallery | about