声明:所有内容来自coursera,作为个人学习笔记记录在这里. 请不要ctrl+c/ctrl+v作业.

Optimization Methods

Until now, you've always used Gradient Descent to update the parameters and minimize the cost. In this notebook, you will learn more advanced optimization methods that can speed up learning and perhaps even get you to a better final value for the cost function. Having a good optimization algorithm can be the difference between waiting days vs. just a few hours to get a good result.

Gradient descent goes "downhill" on a cost function JJ. Think of it as trying to do this:

**Figure 1** : **Minimizing the cost is like finding the lowest point in a hilly landscape**
At each step of the training, you update your parameters following a certain direction to try to get to the lowest possible point.

Notations: As usual, ∂J∂a=∂J∂a= da for any variable a.

To get started, run the following code to import the libraries you will need.

In [35]:

import numpy as np
import matplotlib.pyplot as plt
import scipy.io
import math
import sklearn
import sklearn.datasets from opt_utils import load_params_and_grads, initialize_parameters, forward_propagation, backward_propagation
from opt_utils import compute_cost, predict, predict_dec, plot_decision_boundary, load_dataset
from testCases import * %matplotlib inline
plt.rcParams['figure.figsize'] = (7.0, 4.0) # set default size of plots
plt.rcParams['image.interpolation'] = 'nearest'
plt.rcParams['image.cmap'] = 'gray'
 

1 - Gradient Descent

A simple optimization method in machine learning is gradient descent (GD). When you take gradient steps with respect to all mm examples on each step, it is also called Batch Gradient Descent.

Warm-up exercise: Implement the gradient descent update rule. The gradient descent rule is, for l=1,...,Ll=1,...,L:

W[l]=W[l]−α dW[l](1)(1)W[l]=W[l]−α dW[l]
b[l]=b[l]−α db[l](2)(2)b[l]=b[l]−α db[l]

where L is the number of layers and αα is the learning rate. All parameters should be stored in the parameters dictionary. Note that the iterator l starts at 0 in the for loop while the first parameters are W[1]W[1] and b[1]b[1]. You need to shift l to l+1 when coding.

In [36]:
# GRADED FUNCTION: update_parameters_with_gd

def update_parameters_with_gd(parameters, grads, learning_rate):
"""
Update parameters using one step of gradient descent Arguments:
parameters -- python dictionary containing your parameters to be updated:
parameters['W' + str(l)] = Wl
parameters['b' + str(l)] = bl
grads -- python dictionary containing your gradients to update each parameters:
grads['dW' + str(l)] = dWl
grads['db' + str(l)] = dbl
learning_rate -- the learning rate, scalar. Returns:
parameters -- python dictionary containing your updated parameters
""" L = len(parameters) // 2 # number of layers in the neural networks # Update rule for each parameter
for l in range(L):
### START CODE HERE ### (approx. 2 lines)
parameters["W" + str(l+1)] = parameters["W" + str(l+1)] - learning_rate*grads['dW' + str(l+1)]
parameters["b" + str(l+1)] = parameters["b" + str(l+1)] - learning_rate*grads['db' + str(l+1)]
### END CODE HERE ### return parameters
In [37]:
parameters, grads, learning_rate = update_parameters_with_gd_test_case()

parameters = update_parameters_with_gd(parameters, grads, learning_rate)
print("W1 = " + str(parameters["W1"]))
print("b1 = " + str(parameters["b1"]))
print("W2 = " + str(parameters["W2"]))
print("b2 = " + str(parameters["b2"]))
 
W1 = [[ 1.63535156 -0.62320365 -0.53718766]
[-1.07799357 0.85639907 -2.29470142]]
b1 = [[ 1.74604067]
[-0.75184921]]
W2 = [[ 0.32171798 -0.25467393 1.46902454]
[-2.05617317 -0.31554548 -0.3756023 ]
[ 1.1404819 -1.09976462 -0.1612551 ]]
b2 = [[-0.88020257]
[ 0.02561572]
[ 0.57539477]]
 

Expected Output:

**W1** [[ 1.63535156 -0.62320365 -0.53718766] [-1.07799357 0.85639907 -2.29470142]]
**b1** [[ 1.74604067] [-0.75184921]]
**W2** [[ 0.32171798 -0.25467393 1.46902454] [-2.05617317 -0.31554548 -0.3756023 ] [ 1.1404819 -1.09976462 -0.1612551 ]]
**b2** [[-0.88020257] [ 0.02561572] [ 0.57539477]]
 

A variant of this is Stochastic Gradient Descent (SGD), which is equivalent to mini-batch gradient descent where each mini-batch has just 1 example. The update rule that you have just implemented does not change. What changes is that you would be computing gradients on just one training example at a time, rather than on the whole training set. The code examples below illustrate the difference between stochastic gradient descent and (batch) gradient descent.

  • (Batch) Gradient Descent:
X = data_input
Y = labels
parameters = initialize_parameters(layers_dims)
for i in range(0, num_iterations):
# Forward propagation
a, caches = forward_propagation(X, parameters)
# Compute cost.
cost = compute_cost(a, Y)
# Backward propagation.
grads = backward_propagation(a, caches, parameters)
# Update parameters.
parameters = update_parameters(parameters, grads)
  • Stochastic Gradient Descent:
X = data_input
Y = labels
parameters = initialize_parameters(layers_dims)
for i in range(0, num_iterations):
for j in range(0, m):
# Forward propagation
a, caches = forward_propagation(X[:,j], parameters)
# Compute cost
cost = compute_cost(a, Y[:,j])
# Backward propagation
grads = backward_propagation(a, caches, parameters)
# Update parameters.
parameters = update_parameters(parameters, grads)
 

In Stochastic Gradient Descent, you use only 1 training example before updating the gradients. When the training set is large, SGD can be faster. But the parameters will "oscillate" toward the minimum rather than converge smoothly. Here is an illustration of this:

**Figure 1** : **SGD vs GD**
"+" denotes a minimum of the cost. SGD leads to many oscillations to reach convergence. But each step is a lot faster to compute for SGD than for GD, as it uses only one training example (vs. the whole batch for GD).

Note also that implementing SGD requires 3 for-loops in total:

  1. Over the number of iterations
  2. Over the mm training examples
  3. Over the layers (to update all parameters, from (W[1],b[1])(W[1],b[1]) to (W[L],b[L])(W[L],b[L]))

In practice, you'll often get faster results if you do not use neither the whole training set, nor only one training example, to perform each update. Mini-batch gradient descent uses an intermediate number of examples for each step. With mini-batch gradient descent, you loop over the mini-batches instead of looping over individual training examples.

**Figure 2** : **SGD vs Mini-Batch GD**
"+" denotes a minimum of the cost. Using mini-batches in your optimization algorithm often leads to faster optimization.

What you should remember:

  • The difference between gradient descent, mini-batch gradient descent and stochastic gradient descent is the number of examples you use to perform one update step.
  • You have to tune a learning rate hyperparameter αα.
  • With a well-turned mini-batch size, usually it outperforms either gradient descent or stochastic gradient descent (particularly when the training set is large).
 

2 - Mini-Batch Gradient descent

Let's learn how to build mini-batches from the training set (X, Y).

There are two steps:

  • Shuffle: Create a shuffled version of the training set (X, Y) as shown below. Each column of X and Y represents a training example. Note that the random shuffling is done synchronously between X and Y. Such that after the shuffling the ithith column of X is the example corresponding to the ithith label in Y. The shuffling step ensures that examples will be split randomly into different mini-batches.

  • Partition: Partition the shuffled (X, Y) into mini-batches of size mini_batch_size (here 64). Note that the number of training examples is not always divisible by mini_batch_size. The last mini batch might be smaller, but you don't need to worry about this. When the final mini-batch is smaller than the full mini_batch_size, it will look like this:

Exercise: Implement random_mini_batches. We coded the shuffling part for you. To help you with the partitioning step, we give you the following code that selects the indexes for the 1st1st and 2nd2nd mini-batches:

first_mini_batch_X = shuffled_X[:, 0 : mini_batch_size]
second_mini_batch_X = shuffled_X[:, mini_batch_size : 2 * mini_batch_size]
...

Note that the last mini-batch might end up smaller than mini_batch_size=64. Let ⌊s⌋⌊s⌋ represents ss rounded down to the nearest integer (this is math.floor(s) in Python). If the total number of examples is not a multiple of mini_batch_size=64 then there will be ⌊mmini_batch_size⌋⌊mmini_batch_size⌋ mini-batches with a full 64 examples, and the number of examples in the final mini-batch will be (m−mini_batch_size×⌊mmini_batch_size⌋m−mini_batch_size×⌊mmini_batch_size⌋).

In [38]:
# GRADED FUNCTION: random_mini_batches
def random_mini_batches(X, Y, mini_batch_size = 64, seed = 0):
"""
Creates a list of random minibatches from (X, Y) Arguments:
X -- input data, of shape (input size, number of examples)
Y -- true "label" vector (1 for blue dot / 0 for red dot), of shape (1, number of examples)
mini_batch_size -- size of the mini-batches, integer Returns:
mini_batches -- list of synchronous (mini_batch_X, mini_batch_Y)
""" np.random.seed(seed) # To make your "random" minibatches the same as ours
m = X.shape[1] # number of training examples
mini_batches = [] # Step 1: Shuffle (X, Y)
permutation = list(np.random.permutation(m))
shuffled_X = X[:, permutation]
shuffled_Y = Y[:, permutation].reshape((1,m)) # Step 2: Partition (shuffled_X, shuffled_Y). Minus the end case.
num_complete_minibatches = math.floor(m/mini_batch_size) # number of mini batches of size mini_batch_size in your partitionning
for k in range(0, num_complete_minibatches):
### START CODE HERE ### (approx. 2 lines)
mini_batch_X = shuffled_X[:, k*mini_batch_size : (k+1)*mini_batch_size]
mini_batch_Y = shuffled_Y[:, k*mini_batch_size : (k+1)*mini_batch_size]
### END CODE HERE ###
mini_batch = (mini_batch_X, mini_batch_Y)
mini_batches.append(mini_batch) # Handling the end case (last mini-batch < mini_batch_size)
if m % mini_batch_size != 0:
### START CODE HERE ### (approx. 2 lines)
mini_batch_X = shuffled_X[:, num_complete_minibatches*mini_batch_size : ]
mini_batch_Y = shuffled_Y[:, num_complete_minibatches*mini_batch_size : ]
### END CODE HERE ###
mini_batch = (mini_batch_X, mini_batch_Y)
mini_batches.append(mini_batch) return mini_batches
In [39]:
X_assess, Y_assess, mini_batch_size = random_mini_batches_test_case()
mini_batches = random_mini_batches(X_assess, Y_assess, mini_batch_size) print ("shape of the 1st mini_batch_X: " + str(mini_batches[0][0].shape))
print ("shape of the 2nd mini_batch_X: " + str(mini_batches[1][0].shape))
print ("shape of the 3rd mini_batch_X: " + str(mini_batches[2][0].shape))
print ("shape of the 1st mini_batch_Y: " + str(mini_batches[0][1].shape))
print ("shape of the 2nd mini_batch_Y: " + str(mini_batches[1][1].shape))
print ("shape of the 3rd mini_batch_Y: " + str(mini_batches[2][1].shape))
print ("mini batch sanity check: " + str(mini_batches[0][0][0][0:3]))
shape of the 1st mini_batch_X: (12288, 64)
shape of the 2nd mini_batch_X: (12288, 64)
shape of the 3rd mini_batch_X: (12288, 20)
shape of the 1st mini_batch_Y: (1, 64)
shape of the 2nd mini_batch_Y: (1, 64)
shape of the 3rd mini_batch_Y: (1, 20)
mini batch sanity check: [ 0.90085595 -0.7612069 0.2344157 ]
 

Expected Output:

**shape of the 1st mini_batch_X** (12288, 64)
**shape of the 2nd mini_batch_X** (12288, 64)
**shape of the 3rd mini_batch_X** (12288, 20)
**shape of the 1st mini_batch_Y** (1, 64)
**shape of the 2nd mini_batch_Y** (1, 64)
**shape of the 3rd mini_batch_Y** (1, 20)
**mini batch sanity check** [ 0.90085595 -0.7612069 0.2344157 ]
 

What you should remember:

  • Shuffling and Partitioning are the two steps required to build mini-batches
  • Powers of two are often chosen to be the mini-batch size, e.g., 16, 32, 64, 128.
 

3 - Momentum

Because mini-batch gradient descent makes a parameter update after seeing just a subset of examples, the direction of the update has some variance, and so the path taken by mini-batch gradient descent will "oscillate" toward convergence. Using momentum can reduce these oscillations.

Momentum takes into account the past gradients to smooth out the update. We will store the 'direction' of the previous gradients in the variable vv. Formally, this will be the exponentially weighted average of the gradient on previous steps. You can also think of vv as the "velocity" of a ball rolling downhill, building up speed (and momentum) according to the direction of the gradient/slope of the hill.

**Figure 3**: The red arrows shows the direction taken by one step of mini-batch gradient descent with momentum. The blue points show the direction of the gradient (with respect to the current mini-batch) on each step. Rather than just following the gradient, we let the gradient influence vv and then take a step in the direction of vv.

Exercise: Initialize the velocity. The velocity, vv, is a python dictionary that needs to be initialized with arrays of zeros. Its keys are the same as those in the gradsdictionary, that is: for l=1,...,Ll=1,...,L:

v["dW" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["W" + str(l+1)])
v["db" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["b" + str(l+1)])

Note that the iterator l starts at 0 in the for loop while the first parameters are v["dW1"] and v["db1"] (that's a "one" on the superscript). This is why we are shifting l to l+1 in the for loop.

In [40]:
# GRADED FUNCTION: initialize_velocity

def initialize_velocity(parameters):
"""
Initializes the velocity as a python dictionary with:
- keys: "dW1", "db1", ..., "dWL", "dbL"
- values: numpy arrays of zeros of the same shape as the corresponding gradients/parameters.
Arguments:
parameters -- python dictionary containing your parameters.
parameters['W' + str(l)] = Wl
parameters['b' + str(l)] = bl Returns:
v -- python dictionary containing the current velocity.
v['dW' + str(l)] = velocity of dWl
v['db' + str(l)] = velocity of dbl
""" L = len(parameters) // 2 # number of layers in the neural networks
v = {} # Initialize velocity
for l in range(L):
### START CODE HERE ### (approx. 2 lines)
v["dW" + str(l+1)] = np.zeros((parameters["W" + str(l+1)].shape[0], parameters["W" + str(l+1)].shape[1]))
v["db" + str(l+1)] = np.zeros((parameters["b" + str(l+1)].shape[0], parameters["b" + str(l+1)].shape[1]))
### END CODE HERE ### return v
In [41]:
parameters = initialize_velocity_test_case()

v = initialize_velocity(parameters)
print("v[\"dW1\"] = " + str(v["dW1"]))
print("v[\"db1\"] = " + str(v["db1"]))
print("v[\"dW2\"] = " + str(v["dW2"]))
print("v[\"db2\"] = " + str(v["db2"]))
 
v["dW1"] = [[ 0.  0.  0.]
[ 0. 0. 0.]]
v["db1"] = [[ 0.]
[ 0.]]
v["dW2"] = [[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
v["db2"] = [[ 0.]
[ 0.]
[ 0.]]
 

Expected Output:

**v["dW1"]** [[ 0. 0. 0.] [ 0. 0. 0.]]
**v["db1"]** [[ 0.] [ 0.]]
**v["dW2"]** [[ 0. 0. 0.] [ 0. 0. 0.] [ 0. 0. 0.]]
**v["db2"]** [[ 0.] [ 0.] [ 0.]]
 

Exercise: Now, implement the parameters update with momentum. The momentum update rule is, for l=1,...,Ll=1,...,L:

        

where L is the number of layers, ββ is the momentum and αα is the learning rate. All parameters should be stored in the parameters dictionary. Note that the iterator l starts at 0 in the for loop while the first parameters are W[1]W[1] and b[1]b[1] (that's a "one" on the superscript). So you will need to shift l to l+1 when coding.
In [42]:
# GRADED FUNCTION: update_parameters_with_momentum

def update_parameters_with_momentum(parameters, grads, v, beta, learning_rate):
"""
Update parameters using Momentum Arguments:
parameters -- python dictionary containing your parameters:
parameters['W' + str(l)] = Wl
parameters['b' + str(l)] = bl
grads -- python dictionary containing your gradients for each parameters:
grads['dW' + str(l)] = dWl
grads['db' + str(l)] = dbl
v -- python dictionary containing the current velocity:
v['dW' + str(l)] = ...
v['db' + str(l)] = ...
beta -- the momentum hyperparameter, scalar
learning_rate -- the learning rate, scalar Returns:
parameters -- python dictionary containing your updated parameters
v -- python dictionary containing your updated velocities
""" L = len(parameters) // 2 # number of layers in the neural networks # Momentum update for each parameter
for l in range(L): ### START CODE HERE ### (approx. 4 lines)
# compute velocities
v["dW" + str(l+1)] = beta* v["dW" + str(l+1)]+(1-beta)*grads["dW"+str(l+1)]
v["db" + str(l+1)] = beta* v["db" + str(l+1)]+(1-beta)*grads["db"+str(l+1)]
# update parameters
parameters["W" + str(l+1)] = parameters["W" + str(l+1)]-learning_rate*v["dW" + str(l+1)]
parameters["b" + str(l+1)] = parameters["b" + str(l+1)]-learning_rate*v["db" + str(l+1)]
### END CODE HERE ### return parameters, v
In [43]:
parameters, grads, v = update_parameters_with_momentum_test_case()

parameters, v = update_parameters_with_momentum(parameters, grads, v, beta = 0.9, learning_rate = 0.01)
print("W1 = " + str(parameters["W1"]))
print("b1 = " + str(parameters["b1"]))
print("W2 = " + str(parameters["W2"]))
print("b2 = " + str(parameters["b2"]))
print("v[\"dW1\"] = " + str(v["dW1"]))
print("v[\"db1\"] = " + str(v["db1"]))
print("v[\"dW2\"] = " + str(v["dW2"]))
print("v[\"db2\"] = " + str(v["db2"]))
 
W1 = [[ 1.62544598 -0.61290114 -0.52907334]
[-1.07347112 0.86450677 -2.30085497]]
b1 = [[ 1.74493465]
[-0.76027113]]
W2 = [[ 0.31930698 -0.24990073 1.4627996 ]
[-2.05974396 -0.32173003 -0.38320915]
[ 1.13444069 -1.0998786 -0.1713109 ]]
b2 = [[-0.87809283]
[ 0.04055394]
[ 0.58207317]]
v["dW1"] = [[-0.11006192 0.11447237 0.09015907]
[ 0.05024943 0.09008559 -0.06837279]]
v["db1"] = [[-0.01228902]
[-0.09357694]]
v["dW2"] = [[-0.02678881 0.05303555 -0.06916608]
[-0.03967535 -0.06871727 -0.08452056]
[-0.06712461 -0.00126646 -0.11173103]]
v["db2"] = [[ 0.02344157]
[ 0.16598022]
[ 0.07420442]]
 

Expected Output:

**W1** [[ 1.62544598 -0.61290114 -0.52907334] [-1.07347112 0.86450677 -2.30085497]]
**b1** [[ 1.74493465] [-0.76027113]]
**W2** [[ 0.31930698 -0.24990073 1.4627996 ] [-2.05974396 -0.32173003 -0.38320915] [ 1.13444069 -1.0998786 -0.1713109 ]]
**b2** [[-0.87809283] [ 0.04055394] [ 0.58207317]]
**v["dW1"]** [[-0.11006192 0.11447237 0.09015907] [ 0.05024943 0.09008559 -0.06837279]]
**v["db1"]** [[-0.01228902] [-0.09357694]]
**v["dW2"]** [[-0.02678881 0.05303555 -0.06916608] [-0.03967535 -0.06871727 -0.08452056] [-0.06712461 -0.00126646 -0.11173103]]
**v["db2"]** [[ 0.02344157] [ 0.16598022] [ 0.07420442]]
 

Note that:

  • The velocity is initialized with zeros. So the algorithm will take a few iterations to "build up" velocity and start to take bigger steps.
  • If β=0β=0, then this just becomes standard gradient descent without momentum.

How do you choose ββ?

  • The larger the momentum ββ is, the smoother the update because the more we take the past gradients into account. But if ββ is too big, it could also smooth out the updates too much.
  • Common values for ββ range from 0.8 to 0.999. If you don't feel inclined to tune this, β=0.9β=0.9 is often a reasonable default.
  • Tuning the optimal ββ for your model might need trying several values to see what works best in term of reducing the value of the cost function JJ.
 

What you should remember:

  • Momentum takes past gradients into account to smooth out the steps of gradient descent. It can be applied with batch gradient descent, mini-batch gradient descent or stochastic gradient descent.
  • You have to tune a momentum hyperparameter ββ and a learning rate αα.
 

4 - Adam

Adam is one of the most effective optimization algorithms for training neural networks. It combines ideas from RMSProp (described in lecture) and Momentum.

How does Adam work?

  1. It calculates an exponentially weighted average of past gradients, and stores it in variables vv (before bias correction) and vcorrectedvcorrected (with bias correction).
  2. It calculates an exponentially weighted average of the squares of the past gradients, and stores it in variables ss (before bias correction) and scorrectedscorrected(with bias correction).
  3. It updates parameters in a direction based on combining information from "1" and "2".

The update rule is, for l=1,...,Ll=1,...,L:

⎧⎩⎨⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪vdW[l]=β1vdW[l]+(1−β1)∂J∂W[l]vcorrecteddW[l]=vdW[l]1−(β1)tsdW[l]=β2sdW[l]+(1−β2)(∂J∂W[l])2scorrecteddW[l]=sdW[l]1−(β1)tW[l]=W[l]−αvcorrecteddW[l]scorrecteddW[l]√+ε{vdW[l]=β1vdW[l]+(1−β1)∂J∂W[l]vdW[l]corrected=vdW[l]1−(β1)tsdW[l]=β2sdW[l]+(1−β2)(∂J∂W[l])2sdW[l]corrected=sdW[l]1−(β1)tW[l]=W[l]−αvdW[l]correctedsdW[l]corrected+ε

where:

  • t counts the number of steps taken of Adam
  • L is the number of layers
  • β1β1 and β2β2 are hyperparameters that control the two exponentially weighted averages.
  • αα is the learning rate
  • εε is a very small number to avoid dividing by zero

As usual, we will store all parameters in the parameters dictionary

 

Exercise: Initialize the Adam variables v,sv,s which keep track of the past information.

Instruction: The variables v,sv,s are python dictionaries that need to be initialized with arrays of zeros. Their keys are the same as for grads, that is: for l=1,...,Ll=1,...,L:

v["dW" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["W" + str(l+1)])
v["db" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["b" + str(l+1)])
s["dW" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["W" + str(l+1)])
s["db" + str(l+1)] = ... #(numpy array of zeros with the same shape as parameters["b" + str(l+1)])
In [44]:
# GRADED FUNCTION: initialize_adam

def initialize_adam(parameters) :
"""
Initializes v and s as two python dictionaries with:
- keys: "dW1", "db1", ..., "dWL", "dbL"
- values: numpy arrays of zeros of the same shape as the corresponding gradients/parameters. Arguments:
parameters -- python dictionary containing your parameters.
parameters["W" + str(l)] = Wl
parameters["b" + str(l)] = bl Returns:
v -- python dictionary that will contain the exponentially weighted average of the gradient.
v["dW" + str(l)] = ...
v["db" + str(l)] = ...
s -- python dictionary that will contain the exponentially weighted average of the squared gradient.
s["dW" + str(l)] = ...
s["db" + str(l)] = ... """ L = len(parameters) // 2 # number of layers in the neural networks
v = {}
s = {} # Initialize v, s. Input: "parameters". Outputs: "v, s".
for l in range(L):
### START CODE HERE ### (approx. 4 lines)
v["dW" + str(l+1)] = np.zeros((parameters["W" + str(l+1)].shape[0],parameters["W" + str(l+1)].shape[1]))
v["db" + str(l+1)] = np.zeros((parameters["b" + str(l+1)].shape[0],parameters["b" + str(l+1)].shape[1]))
s["dW" + str(l+1)] = np.zeros((parameters["W" + str(l+1)].shape[0],parameters["W" + str(l+1)].shape[1]))
s["db" + str(l+1)] = np.zeros((parameters["b" + str(l+1)].shape[0],parameters["b" + str(l+1)].shape[1]))
### END CODE HERE ### return v, s
In [45]:
parameters = initialize_adam_test_case()

v, s = initialize_adam(parameters)
print("v[\"dW1\"] = " + str(v["dW1"]))
print("v[\"db1\"] = " + str(v["db1"]))
print("v[\"dW2\"] = " + str(v["dW2"]))
print("v[\"db2\"] = " + str(v["db2"]))
print("s[\"dW1\"] = " + str(s["dW1"]))
print("s[\"db1\"] = " + str(s["db1"]))
print("s[\"dW2\"] = " + str(s["dW2"]))
print("s[\"db2\"] = " + str(s["db2"]))
 
v["dW1"] = [[ 0.  0.  0.]
[ 0. 0. 0.]]
v["db1"] = [[ 0.]
[ 0.]]
v["dW2"] = [[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
v["db2"] = [[ 0.]
[ 0.]
[ 0.]]
s["dW1"] = [[ 0. 0. 0.]
[ 0. 0. 0.]]
s["db1"] = [[ 0.]
[ 0.]]
s["dW2"] = [[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
s["db2"] = [[ 0.]
[ 0.]
[ 0.]]
 

Expected Output:

**v["dW1"]** [[ 0. 0. 0.] [ 0. 0. 0.]]
**v["db1"]** [[ 0.] [ 0.]]
**v["dW2"]** [[ 0. 0. 0.] [ 0. 0. 0.] [ 0. 0. 0.]]
**v["db2"]** [[ 0.] [ 0.] [ 0.]]
**s["dW1"]** [[ 0. 0. 0.] [ 0. 0. 0.]]
**s["db1"]** [[ 0.] [ 0.]]
**s["dW2"]** [[ 0. 0. 0.] [ 0. 0. 0.] [ 0. 0. 0.]]
**s["db2"]** [[ 0.] [ 0.] [ 0.]]
 

Exercise: Now, implement the parameters update with Adam. Recall the general update rule is, for l=1,...,Ll=1,...,L:

⎧⎩⎨⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪vW[l]=β1vW[l]+(1−β1)∂J∂W[l]vcorrectedW[l]=vW[l]1−(β1)tsW[l]=β2sW[l]+(1−β2)(∂J∂W[l])2scorrectedW[l]=sW[l]1−(β2)tW[l]=W[l]−αvcorrectedW[l]scorrectedW[l]√+ε{vW[l]=β1vW[l]+(1−β1)∂J∂W[l]vW[l]corrected=vW[l]1−(β1)tsW[l]=β2sW[l]+(1−β2)(∂J∂W[l])2sW[l]corrected=sW[l]1−(β2)tW[l]=W[l]−αvW[l]correctedsW[l]corrected+ε

Note that the iterator l starts at 0 in the for loop while the first parameters are W[1]W[1] and b[1]b[1]. You need to shift l to l+1 when coding.

In [72]:
# GRADED FUNCTION: update_parameters_with_adam

def update_parameters_with_adam(parameters, grads, v, s, t, learning_rate = 0.01,
beta1 = 0.9, beta2 = 0.999, epsilon = 1e-8):
"""
Update parameters using Adam Arguments:
parameters -- python dictionary containing your parameters:
parameters['W' + str(l)] = Wl
parameters['b' + str(l)] = bl
grads -- python dictionary containing your gradients for each parameters:
grads['dW' + str(l)] = dWl
grads['db' + str(l)] = dbl
v -- Adam variable, moving average of the first gradient, python dictionary
s -- Adam variable, moving average of the squared gradient, python dictionary
learning_rate -- the learning rate, scalar.
beta1 -- Exponential decay hyperparameter for the first moment estimates
beta2 -- Exponential decay hyperparameter for the second moment estimates
epsilon -- hyperparameter preventing division by zero in Adam updates Returns:
parameters -- python dictionary containing your updated parameters
v -- Adam variable, moving average of the first gradient, python dictionary
s -- Adam variable, moving average of the squared gradient, python dictionary
""" L = len(parameters) // 2 # number of layers in the neural networks
v_corrected = {} # Initializing first moment estimate, python dictionary
s_corrected = {} # Initializing second moment estimate, python dictionary # Perform Adam update on all parameters
for l in range(L):
# Moving average of the gradients. Inputs: "v, grads, beta1". Output: "v".
### START CODE HERE ### (approx. 2 lines)
v["dW" + str(l+1)] = beta1*v["dW" + str(l+1)]+(1-beta1)*grads["dW" + str(l+1)]
v["db" + str(l+1)] = beta1*v["db" + str(l+1)]+(1-beta1)*grads["db" + str(l+1)]
### END CODE HERE ### # Compute bias-corrected first moment estimate. Inputs: "v, beta1, t". Output: "v_corrected".
### START CODE HERE ### (approx. 2 lines)
v_corrected["dW" + str(l+1)] = v["dW" + str(l+1)]/(1-beta1**t)
v_corrected["db" + str(l+1)] = v["db" + str(l+1)]/(1-beta1**t)
### END CODE HERE ### # Moving average of the squared gradients. Inputs: "s, grads, beta2". Output: "s".
### START CODE HERE ### (approx. 2 lines)
s["dW" + str(l+1)] = beta2*s["dW" + str(l+1)]+(1-beta2)*(np.power(grads["dW" + str(l+1)],2))
s["db" + str(l+1)] = beta2*s["db" + str(l+1)]+(1-beta2)*(np.power(grads["db" + str(l+1)],2))
### END CODE HERE ### # Compute bias-corrected second raw moment estimate. Inputs: "s, beta2, t". Output: "s_corrected".
### START CODE HERE ### (approx. 2 lines)
s_corrected["dW" + str(l+1)] = s["dW" + str(l+1)]/(1-beta2**t)
s_corrected["db" + str(l+1)] = s["db" + str(l+1)]/(1-beta2**t)
### END CODE HERE ### # Update parameters. Inputs: "parameters, learning_rate, v_corrected, s_corrected, epsilon". Output: "parameters".
### START CODE HERE ### (approx. 2 lines)
parameters["W" + str(l+1)] = parameters["W" + str(l+1)] - learning_rate*v_corrected["dW" + str(l+1)]/(s_corrected["dW" + str(l+1)]**0.5+epsilon)
parameters["b" + str(l+1)] = parameters["b" + str(l+1)] - learning_rate*v_corrected["db" + str(l+1)]/(s_corrected["db" + str(l+1)]**0.5+epsilon)
### END CODE HERE ### return parameters, v, s
In [73]:
parameters, grads, v, s = update_parameters_with_adam_test_case()
parameters, v, s = update_parameters_with_adam(parameters, grads, v, s, t = 2) print("W1 = " + str(parameters["W1"]))
print("b1 = " + str(parameters["b1"]))
print("W2 = " + str(parameters["W2"]))
print("b2 = " + str(parameters["b2"]))
print("v[\"dW1\"] = " + str(v["dW1"]))
print("v[\"db1\"] = " + str(v["db1"]))
print("v[\"dW2\"] = " + str(v["dW2"]))
print("v[\"db2\"] = " + str(v["db2"]))
print("s[\"dW1\"] = " + str(s["dW1"]))
print("s[\"db1\"] = " + str(s["db1"]))
print("s[\"dW2\"] = " + str(s["dW2"]))
print("s[\"db2\"] = " + str(s["db2"]))
 
W1 = [[ 1.63178673 -0.61919778 -0.53561312]
[-1.08040999 0.85796626 -2.29409733]]
b1 = [[ 1.75225313]
[-0.75376553]]
W2 = [[ 0.32648046 -0.25681174 1.46954931]
[-2.05269934 -0.31497584 -0.37661299]
[ 1.14121081 -1.09244991 -0.16498684]]
b2 = [[-0.88529979]
[ 0.03477238]
[ 0.57537385]]
v["dW1"] = [[-0.11006192 0.11447237 0.09015907]
[ 0.05024943 0.09008559 -0.06837279]]
v["db1"] = [[-0.01228902]
[-0.09357694]]
v["dW2"] = [[-0.02678881 0.05303555 -0.06916608]
[-0.03967535 -0.06871727 -0.08452056]
[-0.06712461 -0.00126646 -0.11173103]]
v["db2"] = [[ 0.02344157]
[ 0.16598022]
[ 0.07420442]]
s["dW1"] = [[ 0.00121136 0.00131039 0.00081287]
[ 0.0002525 0.00081154 0.00046748]]
s["db1"] = [[ 1.51020075e-05]
[ 8.75664434e-04]]
s["dW2"] = [[ 7.17640232e-05 2.81276921e-04 4.78394595e-04]
[ 1.57413361e-04 4.72206320e-04 7.14372576e-04]
[ 4.50571368e-04 1.60392066e-07 1.24838242e-03]]
s["db2"] = [[ 5.49507194e-05]
[ 2.75494327e-03]
[ 5.50629536e-04]]
 

Expected Output:

**W1** [[ 1.63178673 -0.61919778 -0.53561312] [-1.08040999 0.85796626 -2.29409733]]
**b1** [[ 1.75225313] [-0.75376553]]
**W2** [[ 0.32648046 -0.25681174 1.46954931] [-2.05269934 -0.31497584 -0.37661299] [ 1.14121081 -1.09245036 -0.16498684]]
**b2** [[-0.88529978] [ 0.03477238] [ 0.57537385]]
**v["dW1"]** [[-0.11006192 0.11447237 0.09015907] [ 0.05024943 0.09008559 -0.06837279]]
**v["db1"]** [[-0.01228902] [-0.09357694]]
**v["dW2"]** [[-0.02678881 0.05303555 -0.06916608] [-0.03967535 -0.06871727 -0.08452056] [-0.06712461 -0.00126646 -0.11173103]]
**v["db2"]** [[ 0.02344157] [ 0.16598022] [ 0.07420442]]
**s["dW1"]** [[ 0.00121136 0.00131039 0.00081287] [ 0.0002525 0.00081154 0.00046748]]
**s["db1"]** [[ 1.51020075e-05] [ 8.75664434e-04]]
**s["dW2"]** [[ 7.17640232e-05 2.81276921e-04 4.78394595e-04] [ 1.57413361e-04 4.72206320e-04 7.14372576e-04] [ 4.50571368e-04 1.60392066e-07 1.24838242e-03]]
**s["db2"]** [[ 5.49507194e-05] [ 2.75494327e-03] [ 5.50629536e-04]]
 

You now have three working optimization algorithms (mini-batch gradient descent, Momentum, Adam). Let's implement a model with each of these optimizers and observe the difference.

 

5 - Model with different optimization algorithms

Lets use the following "moons" dataset to test the different optimization methods. (The dataset is named "moons" because the data from each of the two classes looks a bit like a crescent-shaped moon.)

In [74]:
train_X, train_Y = load_dataset()
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAbcAAAD8CAYAAAD0f+rwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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" alt="" />
 

We have already implemented a 3-layer neural network. You will train it with:

  • Mini-batch Gradient Descent: it will call your function:

    • update_parameters_with_gd()
  • Mini-batch Momentum: it will call your functions:
    • initialize_velocity() and update_parameters_with_momentum()
  • Mini-batch Adam: it will call your functions:
    • initialize_adam() and update_parameters_with_adam()
In [75]:
def model(X, Y, layers_dims, optimizer, learning_rate = 0.0007, mini_batch_size = 64, beta = 0.9,
beta1 = 0.9, beta2 = 0.999, epsilon = 1e-8, num_epochs = 10000, print_cost = True):
"""
3-layer neural network model which can be run in different optimizer modes. Arguments:
X -- input data, of shape (2, number of examples)
Y -- true "label" vector (1 for blue dot / 0 for red dot), of shape (1, number of examples)
layers_dims -- python list, containing the size of each layer
learning_rate -- the learning rate, scalar.
mini_batch_size -- the size of a mini batch
beta -- Momentum hyperparameter
beta1 -- Exponential decay hyperparameter for the past gradients estimates
beta2 -- Exponential decay hyperparameter for the past squared gradients estimates
epsilon -- hyperparameter preventing division by zero in Adam updates
num_epochs -- number of epochs
print_cost -- True to print the cost every 1000 epochs Returns:
parameters -- python dictionary containing your updated parameters
""" L = len(layers_dims) # number of layers in the neural networks
costs = [] # to keep track of the cost
t = 0 # initializing the counter required for Adam update
seed = 10 # For grading purposes, so that your "random" minibatches are the same as ours # Initialize parameters
parameters = initialize_parameters(layers_dims) # Initialize the optimizer
if optimizer == "gd":
pass # no initialization required for gradient descent
elif optimizer == "momentum":
v = initialize_velocity(parameters)
elif optimizer == "adam":
v, s = initialize_adam(parameters) # Optimization loop
for i in range(num_epochs): # Define the random minibatches. We increment the seed to reshuffle differently the dataset after each epoch
seed = seed + 1
minibatches = random_mini_batches(X, Y, mini_batch_size, seed) for minibatch in minibatches: # Select a minibatch
(minibatch_X, minibatch_Y) = minibatch # Forward propagation
a3, caches = forward_propagation(minibatch_X, parameters) # Compute cost
cost = compute_cost(a3, minibatch_Y) # Backward propagation
grads = backward_propagation(minibatch_X, minibatch_Y, caches) # Update parameters
if optimizer == "gd":
parameters = update_parameters_with_gd(parameters, grads, learning_rate)
elif optimizer == "momentum":
parameters, v = update_parameters_with_momentum(parameters, grads, v, beta, learning_rate)
elif optimizer == "adam":
t = t + 1 # Adam counter
parameters, v, s = update_parameters_with_adam(parameters, grads, v, s,
t, learning_rate, beta1, beta2, epsilon) # Print the cost every 1000 epoch
if print_cost and i % 1000 == 0:
print ("Cost after epoch %i: %f" %(i, cost))
if print_cost and i % 100 == 0:
costs.append(cost) # plot the cost
plt.plot(costs)
plt.ylabel('cost')
plt.xlabel('epochs (per 100)')
plt.title("Learning rate = " + str(learning_rate))
plt.show() return parameters
 

You will now run this 3 layer neural network with each of the 3 optimization methods.

5.1 - Mini-batch Gradient descent

Run the following code to see how the model does with mini-batch gradient descent.

In [76]:
# train 3-layer model
layers_dims = [train_X.shape[0], 5, 2, 1]
parameters = model(train_X, train_Y, layers_dims, optimizer = "gd") # Predict
predictions = predict(train_X, train_Y, parameters) # Plot decision boundary
plt.title("Model with Gradient Descent optimization")
axes = plt.gca()
axes.set_xlim([-1.5,2.5])
axes.set_ylim([-1,1.5])
plot_decision_boundary(lambda x: predict_dec(parameters, x.T), train_X, train_Y)
 
Cost after epoch 0: 0.690736
Cost after epoch 1000: 0.685273
Cost after epoch 2000: 0.647072
Cost after epoch 3000: 0.619525
Cost after epoch 4000: 0.576584
Cost after epoch 5000: 0.607243
Cost after epoch 6000: 0.529403
Cost after epoch 7000: 0.460768
Cost after epoch 8000: 0.465586
Cost after epoch 9000: 0.464518
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcMAAAEWCAYAAAAadfxCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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" alt="" />
 
Accuracy: 0.796666666667
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcoAAAEWCAYAAADmYNeIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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%0AioiIhFCiFBERCfH/50bmP+Yj+ssAAAAASUVORK5CYII=" alt="" />
 

5.2 - Mini-batch gradient descent with momentum

Run the following code to see how the model does with momentum. Because this example is relatively simple, the gains from using momemtum are small; but for more complex problems you might see bigger gains.

In [77]:
# train 3-layer model
layers_dims = [train_X.shape[0], 5, 2, 1]
parameters = model(train_X, train_Y, layers_dims, beta = 0.9, optimizer = "momentum") # Predict
predictions = predict(train_X, train_Y, parameters) # Plot decision boundary
plt.title("Model with Momentum optimization")
axes = plt.gca()
axes.set_xlim([-1.5,2.5])
axes.set_ylim([-1,1.5])
plot_decision_boundary(lambda x: predict_dec(parameters, x.T), train_X, train_Y)
 
Cost after epoch 0: 0.690741
Cost after epoch 1000: 0.685341
Cost after epoch 2000: 0.647145
Cost after epoch 3000: 0.619594
Cost after epoch 4000: 0.576665
Cost after epoch 5000: 0.607324
Cost after epoch 6000: 0.529476
Cost after epoch 7000: 0.460936
Cost after epoch 8000: 0.465780
Cost after epoch 9000: 0.464740
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcMAAAEWCAYAAAAadfxCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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%0A+OEAAAAASUVORK5CYII=" alt="" />
 
Accuracy: 0.796666666667
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcoAAAEWCAYAAADmYNeIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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%0AYQAAAABJRU5ErkJggg==" alt="" />
 

5.3 - Mini-batch with Adam mode

Run the following code to see how the model does with Adam.

In [78]:
# train 3-layer model
layers_dims = [train_X.shape[0], 5, 2, 1]
parameters = model(train_X, train_Y, layers_dims, optimizer = "adam") # Predict
predictions = predict(train_X, train_Y, parameters) # Plot decision boundary
plt.title("Model with Adam optimization")
axes = plt.gca()
axes.set_xlim([-1.5,2.5])
axes.set_ylim([-1,1.5])
plot_decision_boundary(lambda x: predict_dec(parameters, x.T), train_X, train_Y)
 
Cost after epoch 0: 0.690552
Cost after epoch 1000: 0.185567
Cost after epoch 2000: 0.150852
Cost after epoch 3000: 0.074454
Cost after epoch 4000: 0.125936
Cost after epoch 5000: 0.104235
Cost after epoch 6000: 0.100552
Cost after epoch 7000: 0.031601
Cost after epoch 8000: 0.111709
Cost after epoch 9000: 0.197648
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAb0AAAEWCAYAAADy9UlpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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%0AMJz0zMysMJz0zMysMJz0zMysMP4P26xojg0XXqQAAAAASUVORK5CYII=" alt="" />
 
Accuracy: 0.94
 
aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcoAAAEWCAYAAADmYNeIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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%0ARLUAAAAASUVORK5CYII=" alt="" />
 

5.4 - Summary

**optimization method** **accuracy** **cost shape**
Gradient descent 79.7% oscillations
Momentum 79.7% oscillations
Adam 94% smoother

Momentum usually helps, but given the small learning rate and the simplistic dataset, its impact is almost negligeable. Also, the huge oscillations you see in the cost come from the fact that some minibatches are more difficult thans others for the optimization algorithm.

Adam on the other hand, clearly outperforms mini-batch gradient descent and Momentum. If you run the model for more epochs on this simple dataset, all three methods will lead to very good results. However, you've seen that Adam converges a lot faster.

Some advantages of Adam include:

  • Relatively low memory requirements (though higher than gradient descent and gradient descent with momentum)
  • Usually works well even with little tuning of hyperparameters (except αα)
 

References:

Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week2, Assignment(Optimization Methods)的更多相关文章

  1. 《Improving Deep Neural Networks:Hyperparameter tuning, Regularization and Optimization》课堂笔记

    Lesson 2 Improving Deep Neural Networks:Hyperparameter tuning, Regularization and Optimization 这篇文章其 ...

  2. [C4] Andrew Ng - Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization

    About this Course This course will teach you the "magic" of getting deep learning to work ...

  3. Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week1, Assignment(Initialization)

    声明:所有内容来自coursera,作为个人学习笔记记录在这里. Initialization Welcome to the first assignment of "Improving D ...

  4. Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week1, Assignment(Gradient Checking)

    声明:所有内容来自coursera,作为个人学习笔记记录在这里. Gradient Checking Welcome to the final assignment for this week! In ...

  5. Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week1, Assignment(Regularization)

    声明:所有内容来自coursera,作为个人学习笔记记录在这里. Regularization Welcome to the second assignment of this week. Deep ...

  6. 课程二(Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization),第一周(Practical aspects of Deep Learning) —— 4.Programming assignments:Gradient Checking

    Gradient Checking Welcome to this week's third programming assignment! You will be implementing grad ...

  7. 吴恩达《深度学习》-课后测验-第二门课 (Improving Deep Neural Networks:Hyperparameter tuning, Regularization and Optimization)-Week 1 - Practical aspects of deep learning(第一周测验 - 深度学习的实践)

    Week 1 Quiz - Practical aspects of deep learning(第一周测验 - 深度学习的实践) \1. If you have 10,000,000 example ...

  8. 吴恩达《深度学习》-第二门课 (Improving Deep Neural Networks:Hyperparameter tuning, Regularization and Optimization)-第一周:深度学习的实践层面 (Practical aspects of Deep Learning) -课程笔记

    第一周:深度学习的实践层面 (Practical aspects of Deep Learning) 1.1 训练,验证,测试集(Train / Dev / Test sets) 创建新应用的过程中, ...

  9. 课程二(Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization),第三周(Hyperparameter tuning, Batch Normalization and Programming Frameworks) —— 2.Programming assignments

    Tensorflow Welcome to the Tensorflow Tutorial! In this notebook you will learn all the basics of Ten ...

随机推荐

  1. Tomcat 一般异常处理方式

    1.启动时出现: Exception starting filter struts2 java.lang.NoClassDefFoundError: org/apache/commons/lang3/ ...

  2. qml: 支持的基本类型

    qml支持的基本类型有: bool unsigned int, int; float    double   qreal QString QUrl QColor QData,  QTime  QDat ...

  3. JAVA8 in Action:行为参数化,匿名类及lambda表达式的初步认知实例整理

    import java.util.ArrayList; import java.util.Arrays; import java.util.List; import java.util.functio ...

  4. Scoring and Modeling—— Underwriting and Loan Approval Process

    https://www.fdic.gov/regulations/examinations/credit_card/ch8.html Types of Scoring FICO Scores    V ...

  5. Hadoop端口说明

    Hadoop端口说明: 默认端口                            设置位置                                    描述信息 8020        ...

  6. Spring Boot笔记六:Thymeleaf介绍

    目录 什么是thymeleaf? 创建最简单的thymeleaf thymeleaf语法 什么是thymeleaf? thymeleaf是一个模板引擎,是用来在Spring Boot中代替JSP的 引 ...

  7. 获取spring上下文的bean 工具类

    有些场景我们不属于controller,service,dao,但是我们需要从spring中得到spring容器里面的bean.这时候我们需要一个类继承 ApplicationContextAware ...

  8. 使用git 上传项目到gitee/github

    参考: https://blog.csdn.net/qq944639839/article/details/79864081 注意:在此之前需要设置ssh公匙 详见:Github/github 初始化 ...

  9. 学习总结:CSS(一)定义方式、选择器、选择器权重

    一.CSS的定义方式 1.内部样式:<style></style> 2.行间样式:<div style="width:100px;height:100px;&q ...

  10. 介绍3款Markdown编辑器

    为什么写此篇  自从CSDN的博客有了Markdown后,慢慢的了解并学会了用Markdown语法写博客.但CSDN博客是在浏览器中使用,于是一直寻找离线的Markdown编辑器.  网上先是找到了M ...