CibelePaulinoAndrade · December 13, 2018 20:13
diff --git a/CoreMLExample.swift b/CoreMLExample.swift
 var model = Flowers()
 guard let prediction = try? model.prediction(data: pixelBuffer!) else {
  return
 }
diff --git a/CoreMLToolsExample.py b/CoreMLToolsExample.py
 #!/usr/bin/env python
 # coding: utf-8

 # In[1]:


 import coremltools


 # In[5]:


 coreml_model = coremltools.converters.caffe.convert(('oxford102.caffemodel', 'deploy.prototxt'), image_input_names='data', class_labels='class_labels.txt')


 # In[4]:


 coreml_model.save('Flowers.mlmodel')


 # In[ ]:
diff --git a/CreateMLExample.swift b/CreateMLExample.swift
 import CreateMLUI

 let builder = MLImageClassifierBuilder()
 builder.showInLiveView()
diff --git a/TuriCreateExample.py b/TuriCreateExample.py
 #!/usr/bin/env python
 # coding: utf-8

 # In[1]:


 import turicreate as tc


 # In[14]:


 data = tc.image_analysis.load_images('Pets-100', with_path=True)


 # In[16]:


 data['label'] = data['path'].apply(lambda path: 'dog' if '/Dog' in path else 'cat')


 # In[17]:


 train_data, test_data = data.random_split(0.8)


 # In[15]:


 model = tc.image_classifier.create(train_data, target = 'label')


 # In[7]:


 predictions = model.predict(test_data)


 # In[8]:


 metrics = model.evaluate(test_data)


 # In[10]:


 metrics['accuracy']


 # In[11]:


 metrics['confusion_matrix']


 # In[12]:


 model.save('mymodel.model')


 # In[13]:


 model.export_coreml('Pets.mlmodel')


 # In[ ]:
	var model = Flowers()
	guard let prediction = try? model.prediction(data: pixelBuffer!) else {
	return
	}
	#!/usr/bin/env python
	# coding: utf-8

	# In[1]:


	import coremltools


	# In[5]:


	coreml_model = coremltools.converters.caffe.convert(('oxford102.caffemodel', 'deploy.prototxt'), image_input_names='data', class_labels='class_labels.txt')


	# In[4]:


	coreml_model.save('Flowers.mlmodel')


	# In[ ]:
	import CreateMLUI

	let builder = MLImageClassifierBuilder()
	builder.showInLiveView()