chore(i18n,curriculum): processed translations (#42868)

This commit is contained in:
camperbot
2021-07-16 11:03:16 +05:30
committed by GitHub
parent 1f5f07cad3
commit 3b32da0191
429 changed files with 7502 additions and 4006 deletions

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da6
title: Conclusion
title: 結論
challengeType: 11
videoId: LMNub5frQi4
dashedName: conclusion
@ -10,19 +10,19 @@ dashedName: conclusion
## --text--
Most people that are experts in AI or machine learning usually...:
大多數人工智能或機器學習專家通常......
## --answers--
have one specialization.
有一個專業。
---
have many specializations.
有很多專業。
---
have a deep understanding of many different frameworks.
對許多不同的框架有深入的瞭解。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d99
title: 'Convolutional Neural Networks: Evaluating the Model'
title: '卷積神經網絡:評估模型'
challengeType: 11
videoId: eCATNvwraXg
dashedName: convolutional-neural-networks-evaluating-the-model
@ -10,19 +10,19 @@ dashedName: convolutional-neural-networks-evaluating-the-model
## --text--
What is **not** a good way to increase the accuracy of a convolutional neural network?
什麼 **不是** 提高卷積神經網絡準確性的好方法?
## --answers--
Augmenting the data you already have.
擴充你已有的數據。
---
Using a pre-trained model.
使用預先訓練的模型。
---
Using your test data to retrain the model.
使用你的測試數據重新訓練模型。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9a
title: 'Convolutional Neural Networks: Picking a Pretrained Model'
title: '卷積神經網絡:選擇預訓練模型'
challengeType: 11
videoId: h1XUt1AgIOI
dashedName: convolutional-neural-networks-picking-a-pretrained-model
@ -10,7 +10,7 @@ dashedName: convolutional-neural-networks-picking-a-pretrained-model
## --text--
Fill in the blanks below to use Google's pre-trained MobileNet V2 model as a base for a convolutional neural network:
填寫下面的空白,使用谷歌預訓練的 MobileNet V2 模型作爲卷積神經網絡的基礎:
```py
base_model = tf.__A__.applications.__B__(input_shape=(160, 160, 3),

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d97
title: 'Convolutional Neural Networks: The Convolutional Layer'
title: '卷積神經網絡:卷積層'
challengeType: 11
videoId: LrdmcQpTyLw
dashedName: convolutional-neural-networks-the-convolutional-layer
@ -10,19 +10,19 @@ dashedName: convolutional-neural-networks-the-convolutional-layer
## --text--
What are the three main properties of each convolutional layer?
每個卷積層的三個主要屬性是什麼?
## --answers--
Input size, the number of filters, and the sample size of the filters.
輸入大小、過濾器數量和過濾器的樣本大小。
---
Input size, input dimensions, and the color values of the input.
輸入大小、輸入尺寸和輸入的顏色值。
---
Input size, input padding, and stride.
輸入大小、輸入填充和步長。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d96
title: Convolutional Neural Networks
title: 卷積神經網絡
challengeType: 11
videoId: _1kTP7uoU9E
dashedName: convolutional-neural-networks
@ -10,19 +10,19 @@ dashedName: convolutional-neural-networks
## --text--
Dense neural networks analyze input on a global scale and recognize patterns in specific areas. Convolutional neural networks...:
密集神經網絡在全局範圍內分析輸入,並識別特定區域的模式。 卷積神經網絡......
## --answers--
also analyze input globally and extract features from specific areas.
也在全局分析輸入並從特定區域提取特徵。
---
do not work well for image classification or object detection.
在圖像分類或物體檢測方面效果不佳。
---
scan through the entire input a little at a time and learn local patterns.
每次一點點地掃描整個輸入,並學習局部模式。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8e
title: 'Core Learning Algorithms: Building the Model'
title: '核心學習算法:構建模型'
challengeType: 11
videoId: 5wHw8BTd2ZQ
dashedName: core-learning-algorithms-building-the-model
@ -10,7 +10,7 @@ dashedName: core-learning-algorithms-building-the-model
## --text--
What kind of estimator/model does TensorFlow recommend using for classification?
TensorFlow 推薦使用哪種估計器/模型進行分類?
## --answers--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8d
title: 'Core Learning Algorithms: Classification'
title: '核心學習算法:分類'
challengeType: 11
videoId: qFF7ZQNvK9E
dashedName: core-learning-algorithms-classification
@ -10,19 +10,19 @@ dashedName: core-learning-algorithms-classification
## --text--
What is classification?
什麼是分類?
## --answers--
The process of separating data points into different classes.
將數據點分離成不同類別的過程。
---
Predicting a numeric value or forecast based on independent and dependent variables.
根據自變量和因變量預測數值或預測。
---
None of the above.
以上都不是。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8f
title: 'Core Learning Algorithms: Clustering'
title: '核心學習算法:聚類'
challengeType: 11
videoId: 8sqIaHc9Cz4
dashedName: core-learning-algorithms-clustering
@ -10,27 +10,27 @@ dashedName: core-learning-algorithms-clustering
## --text--
Which of the following steps is **not** part of the K-Means algorithm?
以下哪個步驟 **不是** K-Means 算法的一部分?
## --answers--
Randomly pick K points to place K centeroids.
隨機選取 K 個點放置 K 個質心。
---
Assign each K point to the closest K centeroid.
將每個 K 點分配給最近的 K 質心。
---
Move each K centeroid into the middle of all of their data points.
將每個 K 質心移動到其所有數據點的中間。
---
Shuffle the K points so they're redistributed randomly.
打亂 K 點,使它們隨機重新分配。
---
Reassign each K point to the closest K centeroid.
重新分配每個 K 點給最近的 K 質心。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d90
title: 'Core Learning Algorithms: Hidden Markov Models'
title: '核心學習算法:隱馬爾可夫模型'
challengeType: 11
videoId: IZg24y4wEPY
dashedName: core-learning-algorithms-hidden-markov-models
@ -10,19 +10,19 @@ dashedName: core-learning-algorithms-hidden-markov-models
## --text--
What makes a Hidden Markov model different than linear regression or classification?
隱馬爾科夫模型與線性迴歸或分類有何不同?
## --answers--
It uses probability distributions to predict future events or states.
它使用概率分佈來預測未來的事件或狀態。
---
It analyzes the relationship between independent and dependent variables to make predictions.
它分析自變量和因變量之間的關係以進行預測。
---
It separates data points into separate categories.
它將數據點分成不同的類別。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8c
title: 'Core Learning Algorithms: The Training Process'
title: '核心學習算法:訓練過程'
challengeType: 11
videoId: _cEwvqVoBhI
dashedName: core-learning-algorithms-the-training-process
@ -10,19 +10,19 @@ dashedName: core-learning-algorithms-the-training-process
## --text--
What are epochs?
什麼是 epoch
## --answers--
The number of times the model will see the same data.
模型看到相同數據的次數。
---
A type of graph.
一種圖。
---
The number of elements you feed to the model at once.
你一次提供給模型的元素數量。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8b
title: 'Core Learning Algorithms: Training and Testing Data'
title: '核心學習算法:訓練和測試數據'
challengeType: 11
videoId: wz9J1slsi7I
dashedName: core-learning-algorithms-training-and-testing-data
@ -10,19 +10,19 @@ dashedName: core-learning-algorithms-training-and-testing-data
## --text--
What is categorical data?
什麼是分類數據?
## --answers--
Another term for one-hot encoding.
獨熱編碼的另一個術語。
---
Any data that is not numeric.
任何非數字的數據。
---
Any data that is represented numerically.
任何以數字表示的數據。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d91
title: 'Core Learning Algorithms: Using Probabilities to make Predictions'
title: '核心學習算法:使用概率進行預測'
challengeType: 11
videoId: fYAYvLUawnc
dashedName: core-learning-algorithms-using-probabilities-to-make-predictions
@ -10,7 +10,7 @@ dashedName: core-learning-algorithms-using-probabilities-to-make-predictions
## --text--
What TensorFlow module should you import to implement `.HiddenMarkovModel()`?
你應該導入什麼 TensorFlow 模塊來實現 `.HiddenMarkovModel()`
## --answers--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d8a
title: 'Core Learning Algorithms: Working with Data'
title: '核心學習算法:處理數據'
challengeType: 11
videoId: u85IOSsJsPI
dashedName: core-learning-algorithms-working-with-data
@ -10,19 +10,19 @@ dashedName: core-learning-algorithms-working-with-data
## --text--
What does the pandas `.head()` function do?
Pandas `.head()` 函數有什麼作用?
## --answers--
Returns the number of entries in a data frame.
返回數據框中的條目數。
---
Returns the number of columns in a data frame.
返回數據框中的列數。
---
By default, shows the first five rows or entries in a data frame.
默認情況下,顯示數據框中的前五行或條目。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d89
title: Core Learning Algorithms
title: 核心學習算法
challengeType: 11
videoId: u5lZURgcWnU
dashedName: core-learning-algorithms
@ -10,25 +10,25 @@ dashedName: core-learning-algorithms
## --text--
Which type of analysis would be best suited for the following problem?:
哪種類型的分析最適合以下問題?
You have the average temperature in the month of March for the last 100 years. Using this data, you want to predict the average temperature in the month of March 5 years from now.
你擁有過去 100 年三月份的平均溫度。 使用此數據,你希望預測 5 年後 3 月的平均溫度。
## --answers--
Multiple regression
多重回歸
---
Correlation
關連
---
Decision tree
決策樹
---
Linear regression
線性迴歸
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d98
title: Creating a Convolutional Neural Network
title: 創建卷積神經網絡
challengeType: 11
videoId: kfv0K8MtkIc
dashedName: creating-a-convolutional-neural-network
@ -10,7 +10,7 @@ dashedName: creating-a-convolutional-neural-network
## --text--
Fill in the blanks below to complete the architecture for a convolutional neural network:
填寫下面的空白以完成卷積神經網絡的架構:
```py
model = models.__A__()

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d87
title: 'Introduction: Machine Learning Fundamentals'
title: '簡介:機器學習基礎'
challengeType: 11
videoId: KwL1qTR5MT8
dashedName: introduction-machine-learning-fundamentals
@ -10,19 +10,19 @@ dashedName: introduction-machine-learning-fundamentals
## --text--
Which statement below is **false**?
以下哪個陳述是 **假的**
## --answers--
Neural networks are modeled after the way the human brain works.
神經網絡以人腦的工作方式爲模型。
---
Computer programs that play tic-tac-toe or chess against human players are examples of simple artificial intelligence.
與人類玩家玩井字棋或國際象棋的計算機程序是簡單人工智能的例子。
---
Machine learning is a subset of artificial intelligence.
機器學習是人工智能的一個子集。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d88
title: Introduction to TensorFlow
title: TensorFlow 簡介
challengeType: 11
videoId: r9hRyGGjOgQ
dashedName: introduction-to-tensorflow
@ -10,11 +10,11 @@ dashedName: introduction-to-tensorflow
## --text--
Which of the following is **not** a type of tensor?
以下哪個 **不是** 張量的類型?
## --answers--
Variable
變量
---
@ -22,15 +22,15 @@ Flowing
---
Placeholder
佔位符
---
SparseTensor
稀疏張量
---
Constant
常量
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da1
title: 'Natural Language Processing With RNNs: Building the Model'
title: '使用 RNN 處理自然語言:構建模型'
challengeType: 11
videoId: 32WBFS7lfsw
dashedName: natural-language-processing-with-rnns-building-the-model
@ -10,7 +10,7 @@ dashedName: natural-language-processing-with-rnns-building-the-model
## --text--
Fill in the blanks below to complete the `build_model` function:
填寫下面的空白以完成 `build_model` 函數:
```py
def build_mode(vocab_size, embedding_dim, rnn_units, batch_size):

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da0
title: 'Natural Language Processing With RNNs: Create a Play Generator'
title: '使用 RNN 進行自然語言處理:創建戲劇生成器'
challengeType: 11
videoId: j5xsxjq_Xk8
dashedName: natural-language-processing-with-rnns-create-a-play-generator
@ -10,7 +10,7 @@ dashedName: natural-language-processing-with-rnns-create-a-play-generator
## --text--
Fill in the blanks below to create the training examples for the RNN:
填寫下面的空白以創建 RNN 的訓練示例:
```py
char_dataset = tf.data.__A__.__B__(text_as_int)
@ -18,7 +18,7 @@ char_dataset = tf.data.__A__.__B__(text_as_int)
## --answers--
A: `DataSet`
A: `Dataset`
B: `from_tensor_slices`

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9f
title: 'Natural Language Processing With RNNs: Making Predictions'
title: '使用 RNN 進行自然語言處理:進行預測'
challengeType: 11
videoId: WO1hINnBj20
dashedName: natural-language-processing-with-rnns-making-predictions
@ -10,19 +10,19 @@ dashedName: natural-language-processing-with-rnns-making-predictions
## --text--
Before you make a prediction with your own review, you should...:
在你通過自己的評論做出預測之前,你應該:
## --answers--
decode the training dataset and compare the results to the test data.
對訓練數據集進行解碼,並將結果與測試數據進行比較。
---
use the encodings from the training dataset to encode your review.
使用訓練數據集的編碼對你的評論進行編碼。
---
assign random values between 0 and the maximum number of vocabulary in your dataset to each word in your review.
將 0 和數據集中最大詞彙量之間的值,隨機分配給你的評論中的每個單詞。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9c
title: 'Natural Language Processing With RNNs: Part 2'
title: '使用 RNN 進行自然語言處理:第 2 部分'
challengeType: 11
videoId: mUU9YXOFbZg
dashedName: natural-language-processing-with-rnns-part-2
@ -10,19 +10,19 @@ dashedName: natural-language-processing-with-rnns-part-2
## --text--
Word embeddings are...:
詞嵌入是指...
## --answers--
an unordered group of encoded words that describes the frequency of words in a given document.
一組無序的編碼詞,描述給定文檔中詞的頻率。
---
a group of encoded words that preserves the original order of the words in a given document.
一組編碼單詞,保留給定文檔中單詞的原始順序。
---
a vectorized representation of words in a given document that places words with similar meanings near each other.
給定文檔中單詞的矢量化表示,將具有相似含義的單詞彼此靠近。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9d
title: 'Natural Language Processing With RNNs: Recurring Neural Networks'
title: '使用 RNN 進行自然語言處理:循環神經網絡'
challengeType: 11
videoId: bX5681NPOcA
dashedName: natural-language-processing-with-rnns-recurring-neural-networks
@ -10,23 +10,23 @@ dashedName: natural-language-processing-with-rnns-recurring-neural-networks
## --text--
What is true about Recurrent Neural Networks?
關於循環神經網絡,哪一項是正確的?
## --answers--
1: They are a type of feed-forward neural network.
1:它們是一種前饋神經網絡。
---
2: They maintain an internal memory/state of the input that was already processed.
2:它們保持着一個已經處理過的輸入的內部存儲器/狀態。
---
3: RNN's contain a loop and process one piece of input at a time.
3RNN 包含一個循環,每次處理一個輸入。
---
4: Both 2 and 3.
42 和 3
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9e
title: 'Natural Language Processing With RNNs: Sentiment Analysis'
title: '使用 RNN 進行自然語言處理:情感分析'
challengeType: 11
videoId: lYeLtu8Nq7c
dashedName: natural-language-processing-with-rnns-sentiment-analysis
@ -10,7 +10,7 @@ dashedName: natural-language-processing-with-rnns-sentiment-analysis
## --text--
Fill in the blanks below to create the model for the RNN:
填寫下面的空白來創建 RNN 模型:
```py
model = __A__.keras.Sequential([

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da2
title: 'Natural Language Processing With RNNs: Training the Model'
title: '使用 RNN 進行自然語言處理:訓練模型'
challengeType: 11
videoId: hEUiK7j9UI8
dashedName: natural-language-processing-with-rnns-training-the-model
@ -10,7 +10,7 @@ dashedName: natural-language-processing-with-rnns-training-the-model
## --text--
Fill in the blanks below to save your model's checkpoints in the `./checkpoints` directory and call the latest checkpoint for training:
填寫下面的空白以將你模型的檢查點保存在 `./checkpoints` 目錄中,並調用最新的檢查點進行訓練:
```py
checkpoint_dir = __A__

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d9b
title: Natural Language Processing With RNNs
title: 使用 RNN 的自然語言處理
challengeType: 11
videoId: ZyCaF5S-lKg
dashedName: natural-language-processing-with-rnns
@ -10,19 +10,19 @@ dashedName: natural-language-processing-with-rnns
## --text--
Natural Language Processing is a branch of artificial intelligence that...:
自然語言處理是人工智能的一個分支......
## --answers--
deals with how computers understand and process natural/human languages.
解決計算機如何理解和處理自然語言/人文語言。
---
translates image data into natural/human languages.
將圖像數據轉換成自然/人文語言。
---
is focused on translating computer languages into natural/human languages.
重點是將計算機語言翻譯成自然/人文語言。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d93
title: 'Neural Networks: Activation Functions'
title: '神經網絡:激活函數'
challengeType: 11
videoId: S45tqW6BqRs
dashedName: neural-networks-activation-functions
@ -10,15 +10,15 @@ dashedName: neural-networks-activation-functions
## --text--
Which activation function switches values between -1 and 1?
哪個激活函數在 -1 和 1 之間切換值?
## --answers--
ReLU (Rectified Linear Unit)
ReLU(線性整流函數)
---
Tanh (Hyperbolic Tangent)
Tanh(雙曲函數)
---

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d95
title: 'Neural Networks: Creating a Model'
title: '神經網絡:創建模型'
challengeType: 11
videoId: K8bz1bmOCTw
dashedName: neural-networks-creating-a-model
@ -10,7 +10,7 @@ dashedName: neural-networks-creating-a-model
## --text--
Fill in the blanks below to build a sequential model of dense layers:
填寫下面的空白,建立一個密集層的順序模型。
```py
model = __A__.__B__([

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d94
title: 'Neural Networks: Optimizers'
title: '神經網絡:優化'
challengeType: 11
videoId: hdOtRPQe1o4
dashedName: neural-networks-optimizers
@ -10,19 +10,19 @@ dashedName: neural-networks-optimizers
## --text--
What is an optimizer function?
什麼是優化函數?
## --answers--
A function that increases the accuracy of a model's predictions.
一種提高模型預測準確性的函數。
---
A function that implements the gradient descent and backpropagation algorithms for you.
一個爲你實現梯度下降和反向傳播算法的函數。
---
A function that reduces the time a model needs to train.
一個可以減少模型訓練時間的函數。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72d92
title: Neural Networks with TensorFlow
title: 使用 TensorFlow 的神經網絡
challengeType: 11
videoId: uisdfrNrZW4
dashedName: neural-networks-with-tensorflow
@ -10,19 +10,19 @@ dashedName: neural-networks-with-tensorflow
## --text--
A densely connected neural network is one in which...:
一個密集連接的神經網絡是這樣一種網絡:
## --answers--
all the neurons in the current layer are connected to one neuron in the previous layer.
當前層中的所有神經元都連接到前一層中的一個神經元。
---
all the neurons in each layer are connected randomly.
每層的所有神經元都是隨機連接的。
---
all the neurons in the current layer are connected to every neuron in the previous layer.
當前層中的所有神經元都連接到前一層中的每個神經元。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da5
title: 'Reinforcement Learning With Q-Learning: Example'
title: '使用 Q-Learning 進行強化學習:示例'
challengeType: 11
videoId: RBBSNta234s
dashedName: reinforcement-learning-with-q-learning-example
@ -10,7 +10,7 @@ dashedName: reinforcement-learning-with-q-learning-example
## --text--
Fill in the blanks to complete the following Q-Learning equation:
填空以完成以下 Q-Learning 方程:
```py
Q[__A__, __B__] = Q[__A__, __B__] + LEARNING_RATE * (reward + GAMMA * np.max(Q[__C__, :]) - Q[__A__, __B__])

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da4
title: 'Reinforcement Learning With Q-Learning: Part 2'
title: '使用 Q-Learning 進行強化學習:第 2 部分'
challengeType: 11
videoId: DX7hJuaUZ7o
dashedName: reinforcement-learning-with-q-learning-part-2
@ -10,15 +10,15 @@ dashedName: reinforcement-learning-with-q-learning-part-2
## --text--
What can happen if the agent does not have a good balance of taking random actions and using learned actions?
如果智能體在採取隨機動作和使用學習動作之間沒有很好的平衡,會發生什麼?
## --answers--
The agent will always try to minimize its reward for the current state/action, leading to local minima.
智能體將始終嘗試將其對當前狀態/動作的獎勵最小化,從而導致局部最小值。
---
The agent will always try to maximize its reward for the current state/action, leading to local maxima.
智能體將始終嘗試將其對當前狀態/動作的獎勵最大化,從而導致局部最大值。
## --video-solution--

View File

@ -1,6 +1,6 @@
---
id: 5e8f2f13c4cdbe86b5c72da3
title: Reinforcement Learning With Q-Learning
title: 使用 Q-Learning 進行強化學習
challengeType: 11
videoId: Cf7DSU0gVb4
dashedName: reinforcement-learning-with-q-learning
@ -10,19 +10,19 @@ dashedName: reinforcement-learning-with-q-learning
## --text--
The key components of reinforcement learning are...
強化學習的關鍵組成部分是......
## --answers--
environment, representative, state, reaction, and reward.
環境、代表、狀態、反應和獎勵。
---
environment, agent, state, action, and reward.
環境、代理、狀態、動作和獎勵。
---
habitat, agent, state, action, and punishment.
環境、代理、狀態、動作和懲罰。
## --video-solution--