def snoopconfig(self, *args, **kwargs):
"""
config snoop by the `snoop,install` method.
see for arguments https://github.com/alexmojaki/snoop/#output-configuration
Args:
*args: see snoop.install
**kwargs: see snoop.install
Returns:
ClsDebugger:the ClsDebugger object (dd)
"""
snoop.install(builtins=False, *args, **kwargs)
return self
def print_everywhere():
"""
https://github.com/alexmojaki/snoop
"""
txt = """
import snoop; snoop.install() ### can be used anywhere
@snoop
def myfun():
from snoop import pp
pp(myvariable)
"""
import snoop
snoop.install() ### can be used anywhere"
print("Decaorator @snoop ")
def main():
foo(1)
snoop.install(enabled=False)
foo(2)
snoop.install(enabled=True)
foo(3)
from pathlib import Path
from my_types import paramsTup
from plot import (multi_bode_axes, multi_bode_add_data, multi_bode_plot_save,
get_json_settings, multi_nyquist_add_data,
multi_nyquist_axes, multi_nyquist_plot_save)
from mung import open_file_numpy, get_params, get_data_numpy, write_imp_data, write_zview_data
import snoop
import pprint
import numpy as np
from collections import OrderedDict
from typing import List, Tuple, Sequence
from dataclasses import dataclass
DEBUG = False
snoop.install(enabled=DEBUG)
@snoop(depth=1)
def main() -> None:
data_dir = str(Path.cwd())
settings = get_json_settings('settings')
both_plot_exists = False
phase_plot_exists = False
gain_plot_exists = False
nyquist_exists = False
voltages = []
for filepath in Path(data_dir).glob('*.csv'):
import pytest
import six
from cheap_repr import cheap_repr, register_repr
from cheap_repr.utils import safe_qualname
from littleutils import file_to_string, string_to_file
from snoop import formatting, install, spy
from snoop.configuration import Config
from snoop.pp_module import is_deep_arg
from snoop.utils import truncate_string, truncate_list, needs_parentheses, NO_ASTTOKENS
current_thread()._ident = current_thread()._Thread__ident = 123456789
formatting._get_filename = lambda _: "/path/to_file.py"
install()
@register_repr(type(cheap_repr))
def repr_function(func, _helper):
return '<function %s at %#x>' % (
safe_qualname(func), id(func))
@register_repr(type(sys))
def repr_module(module, _helper):
return "<module '%s'>" % module.__name__
@register_repr(set)
def repr_set(x, helper):
class Root(DryEnv):
DEBUG = True
SEPARATE_WORKER_PROCESS = False
MASTER_URL = "http://*****:*****@fy0@7c(&lq%)6tt=c+f-(ihd32@t$)i6gjm'
GITHUB_TOKEN = ""
class MONITOR(DryEnv):
ACTIVE = False
THRESHOLD = 90
MIN_PROCESSES = 1
NUM_MEASUREMENTS = 3
SLEEP_TIME = 5
snoop.install(enabled=Root.DEBUG, out=sys.__stderr__, columns=['thread'])
sentry_sdk.init(dsn=Root.SENTRY_DSN,
integrations=[DjangoIntegration()],
send_default_pii=True)
populate_globals()
import torch
from omegaconf import DictConfig, OmegaConf
from transformers import AutoTokenizer
from data import PunctuationDataModule, PunctuationInferenceDataset, PunctuationDomainDatasets
import os
from models import PunctuationDomainModel
from nemo.utils.exp_manager import exp_manager
from time import time
from pytorch_lightning.callbacks import ModelCheckpoint
import atexit
from copy import deepcopy
import snoop
snoop.install()
## 1. Set experiment path here
exp = 'results/2021-03-27_18-00-46'
exp = 'pretrained'
@hydra.main(config_path=f"../Punctuation_with_Domain_discriminator/{exp}/",
config_name="hparams.yaml")
# @hydra.main(config_name="config.yaml")
def main(cfg: DictConfig) -> None:
pl.seed_everything(cfg.seed)
torch.set_printoptions(sci_mode=False)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = PunctuationDomainModel.load_from_checkpoint(
import numpy as np
import snoop
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torchsnooper
from torchsummary import summary
torchsnooper.register_snoop()
snoop.install(enabled=False)
BOARD_SIZE = 11
ACTIONS = list(range(BOARD_SIZE**2))
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
board_ones = torch.ones((BOARD_SIZE, BOARD_SIZE)).to(device)
board_zeros = torch.zeros((BOARD_SIZE, BOARD_SIZE)).to(device)
board_infs = torch.tensor([[float('Inf')] * BOARD_SIZE] *
BOARD_SIZE).to(device)
def action_to_coord(a):
return (a // BOARD_SIZE, a % BOARD_SIZE)
def coord_to_action(i, j):
return i * BOARD_SIZE + j # action index
