Mantéïa

Manteia/ActiveLearning.py

@@ -10,6 +10,7 @@
 import numpy as np
 import math
 from operator import itemgetter
+import random
 
 class RandomSampling():
 	"""
@@ -31,7 +32,7 @@ class UncertaintyEntropySampling():
 	The basic uncertainty sampling query strategy, querying the examples with the top entropy.
 	"""
 
-	def __init__(self,verbose=True):
+	def __init__(self,verbose=False):
 		self.verbose=verbose
 		if self.verbose:
 			print('UncertaintyEntropySampling')

Manteia/Dataset.py

@@ -38,7 +38,7 @@ class Dataset:
 		* **path**        - Path to the data file.
 		
 	"""
-	def __init__(self,name='20newsgroups',train=True,test=False,dev=False,classe=False,desc=False,path='./dataset',verbose=True):
+	def __init__(self,name='20newsgroups',train=True,test=False,dev=False,classe=True,desc=False,path='./dataset',verbose=True):
 		r"""
 		"""
 		self.name=name
@@ -507,23 +507,23 @@ def load_DBPedia(self):
 			download_and_extract(url, self.path)
 
 		self.path_classes = os.path.join(self.path_dir,'classes.txt')
-		classes=['']
+		classes=[]
 		if os.path.isfile(self.path_classes) and self.classe:
 			fi = open(self.path_classes, "r")
 			rows = fi.readlines()
 			for row in rows:
-				classes.append(row.strip())
+					classes.append(row.strip())
 			self.list_labels=classes
-
 		self.path_train = os.path.join(self.path_dir,'train.csv')
 		if os.path.isfile(self.path_train)and self.train:
 			fi = open(self.path_train, "r")
 			rows = fi.readlines()
 			for row in rows:
 				row.strip()
 				row=row.split(',')
+
 				self.documents_train.append(row[1].strip('"')+' '+row[2].strip('"'))
-				self.labels_train.append(classes[int(row[0])])
+				self.labels_train.append(classes[int(row[0])-1])
 		self.path_test = os.path.join(self.path_dir,'test.csv')
 		if os.path.isfile(self.path_test)and self.test:
 			fi = open(self.path_test, "r")
@@ -930,7 +930,7 @@ def load_pubmed_rct20k(self):
 			wget.download(url_test, out=path_dir)
 			wget.download(url_dev, out=path_dir)
 		if self.train:
-			path_file=os.path.join(path_dir,'train.txt')
+			path_file=os.path.join(self.path_dir,'train.txt')
 			fi = open(path_file, "r")
 			rows = fi.readlines()
 			for row in rows:

Manteia/Model.py

@@ -129,7 +129,8 @@ def load_type(self):
 				self.model_type=model_dict[0]
 			else:
 				if self.model_type in model_dict:
-					print('type compatible')
+					if self.verbose:
+						print('type compatible : {}'.format(self.model_type))
 				else:
 					raise TypeError("{} Model type not in : {}".format(self.model_name,model_dict))
 
@@ -302,7 +303,7 @@ def fit(self,train_dataloader,validation_dataloader):
 		#self.model.cuda()
 
 		if self.early_stopping:
-			self.es=EarlyStopping(path=self.path)
+			self.es=EarlyStopping(path=self.path,verbose=self.verbose)
 
 		#if self.n_gpu > 1:
 		#	self.model = torch.nn.DataParallel(self.model)
@@ -361,7 +362,8 @@ def fit(self,train_dataloader,validation_dataloader):
 				avg_train_loss = total_loss / len(train_dataloader)            
 
 				loss_values.append(avg_train_loss)
-				progress(count=step+1, total=len(train_dataloader))
+				if self.verbose:
+					progress(count=step+1, total=len(train_dataloader))
 
 			if self.verbose==True:
 				print("")
@@ -399,7 +401,8 @@ def fit(self,train_dataloader,validation_dataloader):
 						else:tab_logits=np.append(tab_logits,np.argmax(logits, axis=1), axis=0)
 						if tab_labels is None:tab_labels=label_ids
 						else:tab_labels=np.append(tab_labels,label_ids, axis=0)
-					progress(count=step+1, total=len(validation_dataloader))
+					if self.verbose:
+						progress(count=step+1, total=len(validation_dataloader))
 
 
 			acc_validation=accuracy(tab_logits, tab_labels)
@@ -448,7 +451,8 @@ def predict(self,predict_dataloader,p_type='class'):
 
 		self.model.eval()
 		predictions = None
-		print('Predicting :')
+		if self.verbose:
+			print('Predicting :')
 
 		for step,batch in enumerate(predict_dataloader):
 
@@ -474,7 +478,8 @@ def predict(self,predict_dataloader,p_type='class'):
 					if p_type=='probability':
 						if predictions is None:predictions=torch.softmax(logits, dim=1).numpy()
 						else:predictions=np.append(predictions,torch.softmax(logits, dim=1).numpy(), axis=0)
-					progress(count=step+1, total=len(predict_dataloader))
+					if self.verbose:
+						progress(count=step+1, total=len(predict_dataloader))
 
 		return predictions
 
@@ -788,7 +793,8 @@ def save_checkpoint(self, acc_validation, model,device_model):
 		#save by torch
 		device = torch.device('cpu')
 		model.to(device)
-		print(type(model))
+		if self.verbose:
+			print(type(model))
 		#torch.save(model.module.state_dict(),os.path.join(self.path,'state_dict_validation.pt'))
 		torch.save(model.state_dict(),os.path.join(self.path,'state_dict_validation.pt'))
 

Manteia/Visualisation.py

@@ -150,6 +150,42 @@ def plot_train(self,loss,accuracy,granularity=None):
 		if self.show:
 			plt.show()
 
-
-
-
+	def plot_multigraph(self,metric_dictionnary,title=None,labelx=None,labely=None,legend=True):
+
+		x_list=[]
+		y_list=[]
+
+		labely_list=list(metric_dictionnary.keys())
+		for graph in metric_dictionnary.keys():
+			x = []
+			y = []
+			for key,value in metric_dictionnary[graph].items():
+				x.append(float(key))
+				y.append(float(value))
+			x_list.append(x)
+			y_list.append(y)
+
+		tab_color=['red','black','blue','green','orange','purple']
+		plt.axis([0, 1, 0, 1])
+
+		plt.figure(1)
+		i=0
+		for cx,cy,color in zip(x_list,y_list,tab_color):
+			if labely_list[i]=='Base line':
+					courbe=plt.plot(cx, cy,'--',label='Base line',color=color,linewidth=2)
+			else:
+					courbe=plt.plot(cx, cy,label=labely_list[i],color=color,linewidth=2)
+			i=i+1
+		if title is not None:
+			plt.title(title)
+		if labelx is not None:
+			plt.xlabel(labelx)
+		if labely is not None:
+			plt.ylabel(labely)
+		if legend==True:
+			plt.legend(fontsize='large')
+		if self.save:
+			path=os.path.join(self.path,self.name)
+			plt.savefig(path, dpi=300)
+		if self.show:
+			plt.show()

Manteia/__init__.py

@@ -28,7 +28,7 @@
 	reminiscent.
 """
 
-__version__ = "0.0.29"
+__version__ = "0.0.31"
 
 
 from Manteia import Classification

docs/_build/html/ActiveLearning.html

@@ -176,7 +176,7 @@
 
 <dl class="py class">
 <dt id="Manteia.ActiveLearning.UncertaintyEntropySampling">
-<em class="property">class </em><code class="sig-prename descclassname">Manteia.ActiveLearning.</code><code class="sig-name descname">UncertaintyEntropySampling</code><span class="sig-paren">(</span><em class="sig-param"><span class="n">verbose</span><span class="o">=</span><span class="default_value">True</span></em><span class="sig-paren">)</span><a class="headerlink" href="#Manteia.ActiveLearning.UncertaintyEntropySampling" title="Permalink to this definition">¶</a></dt>
+<em class="property">class </em><code class="sig-prename descclassname">Manteia.ActiveLearning.</code><code class="sig-name descname">UncertaintyEntropySampling</code><span class="sig-paren">(</span><em class="sig-param"><span class="n">verbose</span><span class="o">=</span><span class="default_value">False</span></em><span class="sig-paren">)</span><a class="headerlink" href="#Manteia.ActiveLearning.UncertaintyEntropySampling" title="Permalink to this definition">¶</a></dt>
 <dd><p>The basic uncertainty sampling query strategy, querying the examples with the top entropy.</p>
 </dd></dl>
 

docs/_build/html/Dataset.html

@@ -170,7 +170,7 @@
 <span id="dataset"></span><h1>Dataset<a class="headerlink" href="#module-Manteia.Dataset" title="Permalink to this headline">¶</a></h1>
 <span class="target" id="module-Dataset"></span><dl class="py class">
 <dt id="Manteia.Dataset.Dataset">
-<em class="property">class </em><code class="sig-prename descclassname">Manteia.Dataset.</code><code class="sig-name descname">Dataset</code><span class="sig-paren">(</span><em class="sig-param"><span class="n">name</span><span class="o">=</span><span class="default_value">'20newsgroups'</span></em>, <em class="sig-param"><span class="n">train</span><span class="o">=</span><span class="default_value">True</span></em>, <em class="sig-param"><span class="n">test</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">dev</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">classe</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">desc</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">path</span><span class="o">=</span><span class="default_value">'./dataset'</span></em>, <em class="sig-param"><span class="n">verbose</span><span class="o">=</span><span class="default_value">True</span></em><span class="sig-paren">)</span><a class="headerlink" href="#Manteia.Dataset.Dataset" title="Permalink to this definition">¶</a></dt>
+<em class="property">class </em><code class="sig-prename descclassname">Manteia.Dataset.</code><code class="sig-name descname">Dataset</code><span class="sig-paren">(</span><em class="sig-param"><span class="n">name</span><span class="o">=</span><span class="default_value">'20newsgroups'</span></em>, <em class="sig-param"><span class="n">train</span><span class="o">=</span><span class="default_value">True</span></em>, <em class="sig-param"><span class="n">test</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">dev</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">classe</span><span class="o">=</span><span class="default_value">True</span></em>, <em class="sig-param"><span class="n">desc</span><span class="o">=</span><span class="default_value">False</span></em>, <em class="sig-param"><span class="n">path</span><span class="o">=</span><span class="default_value">'./dataset'</span></em>, <em class="sig-param"><span class="n">verbose</span><span class="o">=</span><span class="default_value">True</span></em><span class="sig-paren">)</span><a class="headerlink" href="#Manteia.Dataset.Dataset" title="Permalink to this definition">¶</a></dt>
 <dd><p>This is the class description in order to get some dataset.</p>
 <ul class="simple">
 <li><p><strong>name</strong>        - name of the dataset (str)</p></li>