Add setup instructions for Ubuntu

.gitignore

@@ -10,5 +10,10 @@ examples/checkpoints/*
 examples/chatbot/processed/*
 examples/chatbot/checkpoints/*
 examples/chatbot/data_analysis.py
+examples/data/*
+examples/graphs/*
+examples/test/*
 
 assignments/chatbot/processed/*
+
+env/*

examples/02_feed_dict.py

@@ -19,7 +19,7 @@
 c = a + b  # short for tf.add(a, b)
 
 with tf.Session() as sess:
-	# print(sess.run(c)) # InvalidArgumentError because a doesn’t have any value
+	# print(sess.run(c)) # InvalidArgumentError because a doesnt have any value
 
 	# feed [1, 2, 3] to placeholder a via the dict {a: [1, 2, 3]}
 	# fetch value of c
@@ -35,4 +35,4 @@
 	replace_dict = {a: 15}
 
 	# Run the session, passing in 'replace_dict' as the value to 'feed_dict'
-	print(sess.run(b, feed_dict=replace_dict)) # >> 45
+	print(sess.run(b, feed_dict=replace_dict)) # >> 45

examples/02_simple_tf.py

@@ -16,7 +16,7 @@
 with tf.Session() as sess:
 	writer = tf.summary.FileWriter('./graphs', sess.graph) 
 	print(sess.run(x))
-writer.close() # close the writer when you’re done using it
+writer.close() # close the writer when youre done using it
 
 
 a = tf.constant([2, 2], name='a')
@@ -27,6 +27,7 @@
 # >> [[0 2]
 #	 [4 6]]
 
+shape = [2, 2]
 tf.zeros(shape, dtype=tf.float32, name=None)
 #creates a tensor of shape and all elements will be zeros (when ran in session)
 

examples/03_linear_regression_sol.py

@@ -35,8 +35,8 @@
 Y_predicted = X * w + b 
 
 # Step 5: use the square error as the loss function
-loss = tf.square(Y - Y_predicted, name='loss')
-# loss = utils.huber_loss(Y, Y_predicted)
+#loss = tf.square(Y - Y_predicted, name='loss')
+loss = utils.huber_loss(Y, Y_predicted)
 
 # Step 6: using gradient descent with learning rate of 0.01 to minimize loss
 optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001).minimize(loss)
@@ -48,7 +48,7 @@
 	writer = tf.summary.FileWriter('./graphs/linear_reg', sess.graph)
 
 	# Step 8: train the model
-	for i in range(50): # train the model 100 epochs
+	for i in range(100): # train the model 100 epochs
 		total_loss = 0
 		for x, y in data:
 			# Session runs train_op and fetch values of loss
@@ -67,4 +67,4 @@
 plt.plot(X, Y, 'bo', label='Real data')
 plt.plot(X, X * w + b, 'r', label='Predicted data')
 plt.legend()
-plt.show()
+plt.show()

examples/03_logistic_regression_mnist_sol.py

@@ -20,7 +20,7 @@
 
 # Step 1: Read in data
 # using TF Learn's built in function to load MNIST data to the folder data/mnist
-mnist = input_data.read_data_sets('/data/mnist', one_hot=True) 
+mnist = input_data.read_data_sets('./data/mnist', one_hot=True) 
 
 # Step 2: create placeholders for features and labels
 # each image in the MNIST data is of shape 28*28 = 784
@@ -84,7 +84,7 @@
 	for i in range(n_batches):
 		X_batch, Y_batch = mnist.test.next_batch(batch_size)
 		accuracy_batch = sess.run([accuracy], feed_dict={X: X_batch, Y:Y_batch}) 
-		total_correct_preds += accuracy_batch	
+		total_correct_preds += accuracy_batch[0]
 
 	print('Accuracy {0}'.format(total_correct_preds/mnist.test.num_examples))
 

setup/setup_instruction.md

@@ -62,6 +62,33 @@ $ virtualenv venv --distribute --system-site-packages
 ```
 ### For Ubuntu
 
+Step 1: set up pip and virtual environment
+```bash
+$ sudo apt-get install python-pip
+$ sudo pip install virtualenv
+```
+
+Step 2: set up a project directory. You will do all work for this class in this directory
+```bash
+$ mkdir [my project]
+```
+
+Step 3: set up virtual environment for the project directory. 
+```bash
+$ cd [my project]
+$ virtualenv venv
+```
+These commands create a venv subdirectory in your project where everything is installed.
+
+Step 4: to activate the virtual environment 
+```bash
+$ source venv/bin/activate
+```
+
+Step 5: install tensorflow and other dependencies
+```bash
+$ pip install -r setup/requirements
+```
 
 ### For Windows