Module topicnet.cooking_machine.pretty_output
Expand source code
import numpy as np
from datetime import datetime
from .routine import get_fix_string, get_fix_list
from .routine import get_equal_strings, get_equal_lists
from .models.base_model import MODEL_NAME_LENGTH
UP_END = "┌"
DOWN_END = "└"
MIDDLE = "├"
LAST = "┤"
EMPTY = "│"
START_END = "┐"
SPACE = " "
USELESS_SUBKEYS = {
"type", "verbose", "config"
}
USELESS_KEYS = {
"reuse_theta", "cache_theta", "num_document_passes", "theta_name",
"parent_model_weight", "theta_columns_naming", "transaction_typenames",
"score_tracker", "model_pwt", "model_nwt",
"num_phi_updates", "num_online_processed_batches", "show_progress_bars",
"version", "regularizers",
}
def resize_value(key, value, tab: str = " "):
"""
Parameters
----------
key : str
value : optional
tab: str
(Default value = " ")
Returns
-------
list
"""
if key in ["scores", "taus", "criteria"]:
return [(tab + elem) for elem
in get_fix_list(value, length=-1, num=-1)]
if isinstance(value, (list, tuple, np.ndarray)):
return [(tab + elem) for elem
in get_fix_list(value, length=-1, num=5)]
if isinstance(value, dict):
def _trim(some_value):
if isinstance(some_value, list):
if len(some_value) > 4:
some_value = some_value[:2] + ["..."] + some_value[-2:]
if len(some_value) > 0:
all_strings = all([isinstance(elem, str) for elem in some_value])
if all_strings:
some_value = "[" + ", ".join(some_value) + "]"
return some_value
pairs = [
f"{key}={_trim(some_value)}"
for key, some_value in value.items()
if key not in USELESS_SUBKEYS and some_value is not None
]
return [(tab + elem) for elem in get_fix_list(pairs, length=-1, num=15)]
return [tab + get_fix_string(value, length=-1)]
def get_cube_strings(cubes, tab: str = " ", min_len_per_cube: int = MODEL_NAME_LENGTH):
"""
Parameters
----------
cubes : list of dict
tab : str
(Default value = " ")
min_len_per_cube : int
(Default value = MODEL_NAME_LENGTH defined in BaseModel)
Returns
-------
dict
"""
cube_strings = dict()
for id_cube, cube in enumerate(cubes):
cube_strings[id_cube] = []
cube_strings[id_cube].append(f"{cube['action'].upper()}")
for key, value in cube.items():
if key not in ["action", "params"]:
cube_strings[id_cube] += [get_fix_string(key, length=-1) + ":",
tab + get_fix_string(value, length=-1)]
cube_strings[id_cube].append("")
for params in cube["params"]:
for key, value in params.items():
if (key[0] != "_") and (key[-1] != "_"):
if key not in USELESS_KEYS:
cube_strings[id_cube].append(get_fix_string(key, length=-1) + ":")
cube_strings[id_cube] += resize_value(key, value, tab)
cube_strings[id_cube].append(" + ")
cube_strings[id_cube][-1] = ""
get_equal_strings(cube_strings[id_cube], min_len=min_len_per_cube)
get_equal_lists(cube_strings)
return cube_strings
def get_criteria_strings(criteria, tab: str = " ", min_len_per_cube: int = MODEL_NAME_LENGTH):
"""
Parameters
----------
criteria : list of str
tab : str
(Default value = " ")
min_len_per_cube : int
(Default value = MODEL_NAME_LENGTH defined in BaseModel)
Returns
-------
dict
"""
criterion_strings = dict()
for id_criterion, criterion in enumerate(criteria):
criterion_strings[id_criterion] = []
if criterion is None:
criterion_strings[id_criterion].append(' ')
else:
for statement_id, statement in enumerate(criterion, 1):
stage = statement.split(' and ')
if len(stage) > 1:
heading = f'stage criteria {statement_id}:'
else:
heading = f'stage criterion {statement_id}:'
criterion_strings[id_criterion].append(heading)
for rule in stage:
criterion_strings[id_criterion].append(tab * 2 + rule)
criterion_strings[id_criterion].append("")
get_equal_strings(criterion_strings[id_criterion], min_len=min_len_per_cube)
get_equal_lists(criterion_strings)
return criterion_strings
def add_non_tree_strings(strings, strings_to_add, add_separator=True):
"""
Adding training stage strings
to the experiment description
Parameters
----------
strings : list of strings
description of the experiment
strings_to_add : dict of lists of strings
new information to add to the experiment
add_separator : bool
make pretty separation line
at the end of this strings (Default value = True)
Returns
-------
strings : list of strings
description of the experiment
"""
separation_string = ''
for id_string in range(len(strings_to_add[0])):
string = " "
for id_stage, value in strings_to_add.items():
string += value[id_string] + " | "
if len(separation_string) < len(string):
separation_string += "─" * (len(value[id_string]) + 2) + '+'
string = string[:-3]
strings.append(string)
if add_separator:
strings.append(separation_string[:-3])
return strings
def give_strings_description(experiment,
tab: str = " ",
min_len_per_cube: int = MODEL_NAME_LENGTH,
len_tree_step: int = MODEL_NAME_LENGTH + 1):
"""
Gets strings description of the experiment.
Parameters
----------
tab : str
tab symbol for margin (Default value = " ")
min_len_per_cube : int
minimal length of one stage of description experiment
(Default value = MODEL_NAME_LENGTH defined in BaseModel)
len_tree_step : int
length of the whole one stage description of experiment's tree
(Default value = MODEL_NAME_LENGTH + 1 defined in BaseModel)
Returns
-------
list
strings description
"""
version = 'not defined'
for ind in range(len(experiment.cubes)):
if experiment.cubes[ind]['params'][0].get('version', 'not defined') != 'not defined':
version = experiment.cubes[ind]['params'][0]['version']
strings = [f"Experiment {experiment.experiment_id}", "",
f"Experiment was made with BigARTM {version}"]
cube_strings = get_cube_strings(experiment.cubes, tab, min_len_per_cube)
stage_strings = get_criteria_strings(experiment.criteria, tab, min_len_per_cube)
tree_strings = experiment.tree.get_description()
for key in range(len(cube_strings)):
max_len_string = max(len(cube_strings[key][0]), len(stage_strings[key][0]))
cube_strings[key] = [
get_fix_string(string_cube, length=max_len_string)
for string_cube in cube_strings[key]
]
stage_strings[key] = [
get_fix_string(string_stage, length=max_len_string)
for string_stage in stage_strings[key]
]
# merge strings together
# st = test_len_tree_step - 1
fi = -1
st = len_tree_step - 1
for id_cube, values in cube_strings.items():
fi += len(values[-1]) + 3
for id_string in range(len(tree_strings)):
cur_string = tree_strings[id_string][:]
if st < len(cur_string):
if cur_string[st] == LAST or cur_string[st] == START_END:
tree_strings[id_string] = cur_string[:st] + "─" * (fi - st) \
+ cur_string[st:]
else:
tree_strings[id_string] = cur_string[:st] + " " * (fi - st) \
+ cur_string[st:]
st += fi - st + len_tree_step
strings.append("Tree:")
strings += tree_strings
strings.append("Cubes:")
strings = add_non_tree_strings(strings, cube_strings, add_separator=True)
strings = add_non_tree_strings(strings, stage_strings, add_separator=False)
return strings
def get_html(experiment, window_size: int = 1500):
"""
Gets html text to save human-readable description of the experiment.
Parameters
----------
window_size : int
pixels size of window in html description (Default value = 1500)
Returns
-------
str
description of the experiment in html format
"""
# TODO разобраться с разной шириной пробела в шрифтах
strings = give_strings_description(experiment)
strings_html = ["<html>",
f"<p><font size='+5'>Experiment <b>{experiment.experiment_id}</b></font></p>",
f"<p><i>{strings[2]}</i></p>",
"<p></p>",
f"<td width=\"{window_size}px\" style=\"white-space:pre;\">",
f"<div style=\"width:{window_size}px;overflow:auto;white-space:pre;\">"]
for string in strings[3:]:
if string == "":
strings_html.append("<p></p>")
elif string in ["Tree:", "Cubes:"]:
strings_html.append(f"<p><font size='+1'><b>{string}</b></font></p>")
else:
strings_html += ["<samp><font size='+1'>" + " ".join(string.split(" "))
+ "</font></samp>"]
strings_html += ["</div>", "</td>", "<p></p>",
"<p><i><font size='-1'>Page was generated at "
+ str(datetime.now()) + ".</font></i></p>",
"</html>"]
return "\n".join(strings_html)
def make_notebook_pretty():
from IPython.display import display, HTML
display(HTML("""<style>
div.output_html {
white-space: nowrap;
}
div .output_subarea > pre {
white-space: pre;
word-wrap: normal;
}
div .output_stdout > pre {
white-space: pre-wrap !important;
word-wrap: break-word !important;
}
</style>"""))Functions
def add_non_tree_strings(strings, strings_to_add, add_separator=True)-
Adding training stage strings to the experiment description
Parameters
strings:listofstrings- description of the experiment
strings_to_add:dictoflistsofstrings- new information to add to the experiment
add_separator:bool- make pretty separation line at the end of this strings (Default value = True)
Returns
strings:listofstrings- description of the experiment
Expand source code
def add_non_tree_strings(strings, strings_to_add, add_separator=True): """ Adding training stage strings to the experiment description Parameters ---------- strings : list of strings description of the experiment strings_to_add : dict of lists of strings new information to add to the experiment add_separator : bool make pretty separation line at the end of this strings (Default value = True) Returns ------- strings : list of strings description of the experiment """ separation_string = '' for id_string in range(len(strings_to_add[0])): string = " " for id_stage, value in strings_to_add.items(): string += value[id_string] + " | " if len(separation_string) < len(string): separation_string += "─" * (len(value[id_string]) + 2) + '+' string = string[:-3] strings.append(string) if add_separator: strings.append(separation_string[:-3]) return strings def get_criteria_strings(criteria, tab: str = ' ', min_len_per_cube: int = 26)-
Parameters
criteria:listofstrtab:str- (Default value = " ")
min_len_per_cube:int- (Default value = MODEL_NAME_LENGTH defined in BaseModel)
Returns
dict
Expand source code
def get_criteria_strings(criteria, tab: str = " ", min_len_per_cube: int = MODEL_NAME_LENGTH): """ Parameters ---------- criteria : list of str tab : str (Default value = " ") min_len_per_cube : int (Default value = MODEL_NAME_LENGTH defined in BaseModel) Returns ------- dict """ criterion_strings = dict() for id_criterion, criterion in enumerate(criteria): criterion_strings[id_criterion] = [] if criterion is None: criterion_strings[id_criterion].append(' ') else: for statement_id, statement in enumerate(criterion, 1): stage = statement.split(' and ') if len(stage) > 1: heading = f'stage criteria {statement_id}:' else: heading = f'stage criterion {statement_id}:' criterion_strings[id_criterion].append(heading) for rule in stage: criterion_strings[id_criterion].append(tab * 2 + rule) criterion_strings[id_criterion].append("") get_equal_strings(criterion_strings[id_criterion], min_len=min_len_per_cube) get_equal_lists(criterion_strings) return criterion_strings def get_cube_strings(cubes, tab: str = ' ', min_len_per_cube: int = 26)-
Parameters
cubes:listofdicttab:str- (Default value = " ")
min_len_per_cube:int- (Default value = MODEL_NAME_LENGTH defined in BaseModel)
Returns
dict
Expand source code
def get_cube_strings(cubes, tab: str = " ", min_len_per_cube: int = MODEL_NAME_LENGTH): """ Parameters ---------- cubes : list of dict tab : str (Default value = " ") min_len_per_cube : int (Default value = MODEL_NAME_LENGTH defined in BaseModel) Returns ------- dict """ cube_strings = dict() for id_cube, cube in enumerate(cubes): cube_strings[id_cube] = [] cube_strings[id_cube].append(f"{cube['action'].upper()}") for key, value in cube.items(): if key not in ["action", "params"]: cube_strings[id_cube] += [get_fix_string(key, length=-1) + ":", tab + get_fix_string(value, length=-1)] cube_strings[id_cube].append("") for params in cube["params"]: for key, value in params.items(): if (key[0] != "_") and (key[-1] != "_"): if key not in USELESS_KEYS: cube_strings[id_cube].append(get_fix_string(key, length=-1) + ":") cube_strings[id_cube] += resize_value(key, value, tab) cube_strings[id_cube].append(" + ") cube_strings[id_cube][-1] = "" get_equal_strings(cube_strings[id_cube], min_len=min_len_per_cube) get_equal_lists(cube_strings) return cube_strings def get_html(experiment, window_size: int = 1500)-
Gets html text to save human-readable description of the experiment.
Parameters
window_size:int- pixels size of window in html description (Default value = 1500)
Returns
str- description of the experiment in html format
Expand source code
def get_html(experiment, window_size: int = 1500): """ Gets html text to save human-readable description of the experiment. Parameters ---------- window_size : int pixels size of window in html description (Default value = 1500) Returns ------- str description of the experiment in html format """ # TODO разобраться с разной шириной пробела в шрифтах strings = give_strings_description(experiment) strings_html = ["<html>", f"<p><font size='+5'>Experiment <b>{experiment.experiment_id}</b></font></p>", f"<p><i>{strings[2]}</i></p>", "<p></p>", f"<td width=\"{window_size}px\" style=\"white-space:pre;\">", f"<div style=\"width:{window_size}px;overflow:auto;white-space:pre;\">"] for string in strings[3:]: if string == "": strings_html.append("<p></p>") elif string in ["Tree:", "Cubes:"]: strings_html.append(f"<p><font size='+1'><b>{string}</b></font></p>") else: strings_html += ["<samp><font size='+1'>" + " ".join(string.split(" ")) + "</font></samp>"] strings_html += ["</div>", "</td>", "<p></p>", "<p><i><font size='-1'>Page was generated at " + str(datetime.now()) + ".</font></i></p>", "</html>"] return "\n".join(strings_html) def give_strings_description(experiment, tab: str = ' ', min_len_per_cube: int = 26, len_tree_step: int = 27)-
Gets strings description of the experiment.
Parameters
tab:str- tab symbol for margin (Default value = " ")
min_len_per_cube:int- minimal length of one stage of description experiment (Default value = MODEL_NAME_LENGTH defined in BaseModel)
len_tree_step:int- length of the whole one stage description of experiment's tree (Default value = MODEL_NAME_LENGTH + 1 defined in BaseModel)
Returns
list- strings description
Expand source code
def give_strings_description(experiment, tab: str = " ", min_len_per_cube: int = MODEL_NAME_LENGTH, len_tree_step: int = MODEL_NAME_LENGTH + 1): """ Gets strings description of the experiment. Parameters ---------- tab : str tab symbol for margin (Default value = " ") min_len_per_cube : int minimal length of one stage of description experiment (Default value = MODEL_NAME_LENGTH defined in BaseModel) len_tree_step : int length of the whole one stage description of experiment's tree (Default value = MODEL_NAME_LENGTH + 1 defined in BaseModel) Returns ------- list strings description """ version = 'not defined' for ind in range(len(experiment.cubes)): if experiment.cubes[ind]['params'][0].get('version', 'not defined') != 'not defined': version = experiment.cubes[ind]['params'][0]['version'] strings = [f"Experiment {experiment.experiment_id}", "", f"Experiment was made with BigARTM {version}"] cube_strings = get_cube_strings(experiment.cubes, tab, min_len_per_cube) stage_strings = get_criteria_strings(experiment.criteria, tab, min_len_per_cube) tree_strings = experiment.tree.get_description() for key in range(len(cube_strings)): max_len_string = max(len(cube_strings[key][0]), len(stage_strings[key][0])) cube_strings[key] = [ get_fix_string(string_cube, length=max_len_string) for string_cube in cube_strings[key] ] stage_strings[key] = [ get_fix_string(string_stage, length=max_len_string) for string_stage in stage_strings[key] ] # merge strings together # st = test_len_tree_step - 1 fi = -1 st = len_tree_step - 1 for id_cube, values in cube_strings.items(): fi += len(values[-1]) + 3 for id_string in range(len(tree_strings)): cur_string = tree_strings[id_string][:] if st < len(cur_string): if cur_string[st] == LAST or cur_string[st] == START_END: tree_strings[id_string] = cur_string[:st] + "─" * (fi - st) \ + cur_string[st:] else: tree_strings[id_string] = cur_string[:st] + " " * (fi - st) \ + cur_string[st:] st += fi - st + len_tree_step strings.append("Tree:") strings += tree_strings strings.append("Cubes:") strings = add_non_tree_strings(strings, cube_strings, add_separator=True) strings = add_non_tree_strings(strings, stage_strings, add_separator=False) return strings def make_notebook_pretty()-
Expand source code
def make_notebook_pretty(): from IPython.display import display, HTML display(HTML("""<style> div.output_html { white-space: nowrap; } div .output_subarea > pre { white-space: pre; word-wrap: normal; } div .output_stdout > pre { white-space: pre-wrap !important; word-wrap: break-word !important; } </style>""")) def resize_value(key, value, tab: str = ' ')-
Parameters
key:strvalue:optionaltab:str- (Default value = " ")
Returns
list
Expand source code
def resize_value(key, value, tab: str = " "): """ Parameters ---------- key : str value : optional tab: str (Default value = " ") Returns ------- list """ if key in ["scores", "taus", "criteria"]: return [(tab + elem) for elem in get_fix_list(value, length=-1, num=-1)] if isinstance(value, (list, tuple, np.ndarray)): return [(tab + elem) for elem in get_fix_list(value, length=-1, num=5)] if isinstance(value, dict): def _trim(some_value): if isinstance(some_value, list): if len(some_value) > 4: some_value = some_value[:2] + ["..."] + some_value[-2:] if len(some_value) > 0: all_strings = all([isinstance(elem, str) for elem in some_value]) if all_strings: some_value = "[" + ", ".join(some_value) + "]" return some_value pairs = [ f"{key}={_trim(some_value)}" for key, some_value in value.items() if key not in USELESS_SUBKEYS and some_value is not None ] return [(tab + elem) for elem in get_fix_list(pairs, length=-1, num=15)] return [tab + get_fix_string(value, length=-1)]