def _maybe_import_r_lib(name: str,
raise_exc: bool = False) -> Optional[Tuple[Any, str]]:
global _r_lib, _r_lib_name
if name == _r_lib_name and _r_lib is not None:
return _r_lib, _r_lib_name
try:
from logging import ERROR
from rpy2.robjects import r
from rpy2.robjects.packages import PackageNotInstalledError, importr
from rpy2.rinterface_lib.callbacks import logger
logger.setLevel(ERROR)
r["options"](warn=-1)
except ImportError as e:
raise ImportError(
"Unable to import `rpy2`, install it first as `pip install rpy2` version `>=3.3.0`."
) from e
try:
_r_lib = importr(name)
_r_lib_name = name
return _r_lib, _r_lib_name
except PackageNotInstalledError as e:
if not raise_exc:
return
raise RuntimeError(
f"Install R library `{name!r}` first as `install.packages({name!r}).`"
) from e
def dirichletreg_df(prop_df,
covar_df,
formula,
onevsrest_category=None,
return_reg_input=False):
from rpy2.robjects import r, Formula
from rpy2.robjects.packages import importr
from rpy2.rinterface_lib.callbacks import logger as rpy2_logger
dr = importr('DirichletReg')
dr_df = pd.concat([prop_df, covar_df], axis=1)
f = Formula(formula)
rpy2_logger.setLevel(
logging.ERROR) # will display errors, but not warnings
f.environment['y'] = dr.DR_data(py2r(prop_df))
rpy2_logger.setLevel(
logging.WARNING) # will display errors, but not warnings
if onevsrest_category is None:
fit = dr.DirichReg(f, py2r(dr_df))
else:
assert onevsrest_category in prop_df.columns
cat_index = prop_df.columns.tolist().index(onevsrest_category) + 1
fit = dr.DirichReg(f,
py2r(dr_df),
model='alternative',
**{'sub.comp': cat_index})
r.sink(file='/dev/null')
u = r.summary(fit)
r.sink()
if r('sink.number')()[0] > 0:
r.sink()
if onevsrest_category is None:
varnames = u.rx2('varnames')
else:
varnames = [onevsrest_category] * 2
coef_mat = u.rx2('coef.mat')
rows = r2py(r('rownames')(coef_mat))
coef_df = r2py(r('as.data.frame')(coef_mat)).reset_index(drop=True)
coef_df.columns = ['coefficient', 'se', 'zval', 'pval']
coef_df['compartment'] = np.repeat(varnames, r2py(u.rx2('n.vars')))
coef_df['variable'] = rows
coef_df['significance'] = bin_pval(coef_df.pval)
if onevsrest_category is not None:
coef_df['coef_type'] = np.repeat(['mean', 'precision'],
r2py(u.rx2('n.vars')))
if return_reg_input:
return dr_df, coef_df
else:
return coef_df
from rpy2.robjects import r
from rpy2.rinterface_lib.callbacks import logger as rpy2_logger
import logging
rpy2_logger.setLevel(logging.ERROR)
# rpy2_logger.addFilter(lambda record: 'Model is being refit with current smoothing parameters but initial states are being re-estimated' not in record.msg)
r("library(forecast)")
auto_arima = r(
"""
function(y, xreg, seasonal){
ts(y) %>% auto.arima(max.p=10, max.q=3, max.order=10, xreg=xreg, seasonal=seasonal, lambda='auto', stepwise=F, approximation=F)
}
"""
)
get_arima_order = r("function(model){ model %>% arimaorder()}")
get_arima_lambda = r("function(model){ model$lambda}")
fit_arima = r(
"""
function(order, y_train, X_train){
ts(y_train) %>%
Arima(order=order, xreg=X_train)
}
"""
)
refit_arima = r(
"""
function(model, y_train, X_train){
def read_rFCS(file_path):
from rpy2.robjects import globalenv, pandas2ri, r
from rpy2.robjects.packages import importr
from rpy2.rinterface_lib.callbacks import logger
import logging
logger.setLevel(logging.ERROR)
pandas2ri.activate()
flowcore = importr("flowCore")
base = importr("base")
arg_transf = "transformation=FALSE"
raw_FCS = flowcore.read_FCS(str(file_path), transformation=False,)
r('''
marker_names<-function(FF){
return(flowCore::markernames(FF))
}
FF2dframe<-function(FF){
if(class(FF) == "flowFrame"){
return(as.data.frame(exprs(FF)))}
else {
stop("Object is not of type flowFrame")}
}
''')
fcs_columns = globalenv["marker_names"](raw_FCS)
df_file = globalenv["FF2dframe"](raw_FCS)
if not isinstance(fcs_columns,np.ndarray): #Error checking for StrVector output from rpy2
#This particular issue appeared with an update to a dependency in 20Q4/21Q1
columns_list = []
for i in fcs_columns:
columns_list.append(i)
fcs_columns = columns_list
else:
fcs_columns = fcs_columns.tolist()
if not isinstance(df_file,pd.core.frame.DataFrame): #Error if the above can't be solved
print("CALLUM'S ERROR. CONTACT @FerranC96")
print(type(df_file),": ", df_file)
for i in df_file:
print (i)
original_columns = df_file.columns.values.tolist()
filtered_columns = []
try:
reg_markers = re.compile("(\d+Di$)")#|_dist$)") #All isotopes + debarcoder info
reg_filter = re.compile("^\d+[A-Za-z]+")
match_list=[]
for i in original_columns:
if reg_markers.search(i):
match_list.append(i)
for i in fcs_columns:
if reg_filter.search(i):
filtered_columns.append(i)
if len(filtered_columns)==len(match_list) and len(filtered_columns)!=0:
for new_n,old_n in zip(filtered_columns,match_list):
df_file.rename({old_n: new_n}, axis="columns", inplace=True)
no_filter=False
elif len(match_list)==0 and len(filtered_columns)!=0: #Columns already have PnS?
preprocessed_cols = []
for i in original_columns:
preprocessed_cols.append(i)
if preprocessed_cols==filtered_columns: #Indeed, PnS and "PnN" match!
print("WARNING: FCS was already processed. Keeping original columns")
no_filter=False
else:
raise ValueError
else:
print("WARNING: $PnS (desc in flowCore) and $PnN channels don't match")
print("marker PnS",fcs_columns)
print("marker PnN (original column names)", match_list)
raise ValueError
except ValueError:
print("ERROR: Couldn't read $PnS channel names")
print("No filtering will be performed. Please manually rename your channels")
no_filter=True #Not applying filtering to avoid losing unwated cols
return df_file, no_filter
from rpy2.rinterface_lib.callbacks import logger as rpy2_logger
import logging
rpy2_logger.setLevel(logging.ERROR)
'''Beginning with 3.0 version, rpy2 does not use Python's warnings module anymore.
It now relies on logging instead. The new solution would be:
from rpy2.rinterface_lib.callbacks import logger as rpy2_logger
import logging
rpy2_logger.setLevel(logging.ERROR) # will display errors, but not warnings
If you want to filter specific warnings only, use:
rpy2_logger.addFilter(lambda record: 'notch went outside hinges' not in record.msg)
See LogRecord class specification for available fields for advanced filtering.
'''
#import warnings
#from rpy2.rinterface import RRuntimeWarning
#warnings.filterwarnings("ignore", category=RRuntimeWarning)
import rpy2.robjects as robjects
from rpy2.robjects import r
import rpy2.robjects.numpy2ri
robjects.numpy2ri.activate()
robjects.r("library(baseline)")
import matplotlib.pyplot as plt
import numpy as np
from scipy import stats
t = np.linspace(0, 100, 101)
rate = 10 * stats.norm(50, 1).pdf(t) + np.random.rand(101)
from statsmodels.tsa.stattools import adfuller
from rpy2.robjects.packages import importr
import rpy2.robjects as robjects
from rpy2.rinterface_lib.callbacks import logger as rpy2_logger
import logging
rpy2_logger.setLevel(logging.ERROR) # will display errors, but not warnings
import bayesian_changepoint_detection.online_changepoint_detection as oncd
from functools import partial
import matplotlib.cm as cm
import matplotlib.pyplot as plt
import numpy as np
import copy
import pandas as pd
from arch.unitroot import PhillipsPerron, KPSS
# # for use with cox stuart test in R
# webr = importr('webr')
# cox_stuart_test = robjects.r('cox.stuart.test')
# for use with findfrequency in R
forecast = importr("forecast")
findfrequency = robjects.r('findfrequency')
def concept_drift(vals, plot=False, verbose=False):
# https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.pcolor.html#matplotlib.pyplot.pcolor
# https://stackoverflow.com/questions/42687454/pcolor-data-plot-in-python