0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

機械学習英語

0
Posted at

誰が-する[誰に][何を](修飾語)(場所)(時)

The engineer trained the model using clean labeled data in the cloud environment yesterday.
(エンジニアが昨日クラウド環境でクリーンなラベル付きデータを使ってモデルを訓練した)

The researcher presented the results to the team during the weekly meeting.
(研究者が週次ミーティングでチームに結果を発表した)

The AI system recommended personalized content to the user based on their activity history.
(AIシステムがユーザーの行動履歴に基づきパーソナライズされたコンテンツを推薦した)

誰がする・だ[誰を・何を][どのような](修飾語)(場所)

The model predicted future sales trends with high accuracy on real-world datasets.
(モデルが実データ上で高精度に将来の売上傾向を予測した)

The algorithm classified the images with impressive speed on a GPU server.
(アルゴリズムがGPUサーバ上で画像を高速分類した)

The assistant analyzed the sentiment of customer reviews using a pre-trained BERT model.
(アシスタントが事前学習済みBERTモデルを使って顧客レビューの感情を分析した)

Chapter 0: Parts of Speech in Machine Learning Context

★ Noun: names of people, things, or concepts (e.g., model, dataset, algorithm)
★ Verb: actions or states (e.g., train, learn, predict)
★ Adjective: describes a noun (e.g., accurate model)
★ Adverb: modifies verbs or adjectives (e.g., efficiently trained)


Question 1: Choose ① for countable nouns, ② for uncountable nouns.

model → [①]
data → [②]
algorithm → [①]
knowledge → [②]
dataset → [①]
noise → [②]
result → [①]


Question 2: Choose the part of speech for each word.
① Noun ② Verb ③ Adjective ④ Adverb

train → [②]
accurate → [③]
input → [①]
efficiently → [④]
analyze → [②]
experiment → [①]
useful → [③]
independently → [④]
learning → [①]
deploy → [②]

Lesson 1: be動詞と一般動詞(機械学習バージョン)
空所に入る最も適切な語の番号を選べ。

【問題1】
The model (  ) very accurate now. It needs more data for improvement.
1. am
2. is
3. are
4. was
5. were
→ 答え: 2

【問題2】
My team (  ) the algorithm to us last week.
1. show
2. shows
3. showed
4. is
5. was
→ 答え: 3

【問題3】
There (  ) a major bug in the training code last night.
1. is
2. is
3. are
4. was
5. were
→ 答え: 4

【問題4】
We (  ) Python and TensorFlow every day.
1. was
2. is
3. study
4. studies
5. studied
→ 答え: 3

【問題5】
Dr. Smith (  ) us deep learning last semester.
1. teach
2. teaches
3. teached
4. taught
5. be
→ 答え: 4
Lesson 2: 否定文・疑問文(機械学習バージョン)
空所に入るもっとも適切な語句の番号を選べ。

【問題1】
Does the AI model perform well on test data?
— Yes, it (  ).
1. is
2. was
3. does
4. did
→ 答え: 3

【問題2】
The developer (  ) train the model yesterday.
1. didn’t train
2. didn’t trained
3. wasn’t train
4. wasn’t trained
→ 答え: 1

【問題3】
(  ) you working on the data pipeline last night?
1. Did
2. Do
3. Were
4. Are
→ 答え: 3

【問題4】
I (  ) a machine learning engineer now.
1. don’t
2. didn’t
3. am not
4. wasn’t
5. amn’t
→ 答え: 3

【問題5】
(  ) this script preprocess the data correctly?
1. Do
2. Does
3. Am
4. Is
5. Are
→ 答え: 2
(機械学習をテーマとした第3講:文型チェック問題)。

【選択肢】
① 第1文型(SV)
② 第2文型(SVC)
③ 第3文型(SVO)
④ 第4文型(SVOO)
⑤ 第5文型(SVOC)

【問題】次の英文の文型を番号で答えなさい。
1. The model converged quickly.
2. The results seem accurate.
3. The engineer trained the model.
4. The professor gave the student a dataset.
5. We consider this algorithm efficient.

【解答】
1. ①
2. ②
3. ③
4. ④
5. ⑤

CHECK問題(第4講)のプレーンテキスト版です。

【文法まとめ】
★ 現在形:現在も継続的に成り立つ動作や状態
★ 過去形:過去に完了した事実や行為
★ 大過去形:過去よりもさらに過去の動作
★ 未来形:これから起こること
★ 進行形:動作が進行中であることを強調

【選択肢】
1. 現在形
2. 過去形
3. 大過去形
4. 未来形
5. 進行形

【機械学習をテーマにした問題】
空所に最も適する語句を(番号)で選べ。

1.	Our model (    ) ready for deployment tomorrow.

 1 will
 2 is
 3 am going
 4 will be
→ 答え:4

2.	The data scientist (    ) a report when I entered the lab.

 1 made
 2 was making
 3 making
 4 is made
→ 答え:2

3.	The system (    ) a large dataset every morning.

 1 is having
 2 has
 3 having
 4 have
→ 答え:2

4.	The engineer discovered that the model (    ) the test accuracy.

 1 lost
 2 loses
 3 had lost
 4 was losing
→ 答え:3

5.	The student (    ) about gradient descent last night.

 1 studied
 2 will study
 3 studies
 4 had studied
→ 答え:1

機械学習例文(進行形にしない動詞)
1. think(考える)
 I think this model is overfitting the training data.
 (このモデルは学習データに過学習していると思う。)
2. like(好む)
 Most researchers like using PyTorch for custom models.
 (多くの研究者はカスタムモデルにPyTorchを使うのを好む。)
3. hope(望む)
 We hope the next iteration will improve accuracy.
 (次の反復で精度が向上することを願っている。)
4. hear(聞こえる)
 I hear that the latest model from OpenAI performs well on benchmarks.
 (OpenAIの最新モデルはベンチマークで高性能だと聞いている。)
5. see(見える)
 We see a clear improvement in the loss curve.
 (損失曲線に明らかな改善が見える。)
6. believe(信じる)
 They believe their new algorithm can outperform the baseline.
 (彼らは新しいアルゴリズムがベースラインを上回れると信じている。)
7. have(持っている)
 The model has over 10 million parameters.
 (そのモデルは1000万以上のパラメータを持っている。)
8. belong to(属している)
 This dataset belongs to the UCI Machine Learning Repository.
 (このデータセットはUCI機械学習リポジトリに属している。)
9. resemble(似ている)
 This neural architecture resembles a Transformer but with fewer layers.
 (このニューラルアーキテクチャはTransformerに似ているが、層数が少ない。)
10. depend on(依存する)
 Model performance depends on the quality of the input features.
 (モデル性能は入力特徴量の質に依存する。)

0
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?