class Data(object):
"""
Keeps the data setup that it will pass to the gui and the gui will then update and interact with it
"""
def __init__(self):
"""
Sets up the data
"""
self._y=[]
self._dic_of_names={}
self._names_int={}
self._names_count=0
self._X=[]
self._clf=Clf().clf
def get_names(self):
"""
Returns a list of all the current labeled names besides the Unknown value.
"""
lst = self._names_int.keys()
if 'Unknown' in lst:
lst.remove('Unknown')
if len(lst)>2:
lst=sorted(lst)
return lst
def add_data(self,X,y,path):
"""
If the model is fit then it will return the best predicted guess
"""
##add tags on the file
add_attr_to_file(y,path)
##keeps the currect name_dict
if y not in self._names_int:
self._dic_of_names[self._names_count]=y
self._names_int[y]=self._names_count
self._names_count +=1
self._y.append(self._names_int[y])
self._X.append(X)
##The n of the pca and lda n will adjust based off of size of training data
##It will train after size of 15 and retain after every 5 new inputs
if len(self._y)%5==0 and self._names_count>1 and len(self._y)>15:
self._clf=Clf(
pca_n=100,
lda_n=len(set(self._y))
).clf.fit(self._X,self._y)
def get_best_guess(self,X):
"""
If the model is fit then it will return the best predicted guess
"""
try:
return self._dic_of_names[self._clf.predict(X)[0]]
except NotFittedError:
pass
return None
def __init__(self):
"""
Sets up the data
"""
self._y=[]
self._dic_of_names={}
self._names_int={}
self._names_count=0
self._X=[]
self._clf=Clf().clf
def add_data(self,X,y,path):
"""
If the model is fit then it will return the best predicted guess
"""
##add tags on the file
add_attr_to_file(y,path)
##keeps the currect name_dict
if y not in self._names_int:
self._dic_of_names[self._names_count]=y
self._names_int[y]=self._names_count
self._names_count +=1
self._y.append(self._names_int[y])
self._X.append(X)
##The n of the pca and lda n will adjust based off of size of training data
##It will train after size of 15 and retain after every 5 new inputs
if len(self._y)%5==0 and self._names_count>1 and len(self._y)>15:
self._clf=Clf(
pca_n=100,
lda_n=len(set(self._y))
).clf.fit(self._X,self._y)
