Skip to content
Home » [Update] Machine Learning บทที่ 5: Categorical Encoding | shape แปลว่า – NATAVIGUIDES

[Update] Machine Learning บทที่ 5: Categorical Encoding | shape แปลว่า – NATAVIGUIDES

shape แปลว่า: คุณกำลังดูกระทู้

Categorical Encoding

โดย ชิตพงษ์ กิตตินราดร | ธันวาคม 2562

ในบทนี้เราจะเริ่มมาเจาะลึกในรายละเอียด โดยจะอธิบายเรื่องการเตรียมข้อมูลให้มีความพร้อมสำหรับการสร้างโมเดล

เราได้ลองสร้าง Machine learning model มาสองเรื่องแล้ว ถ้าสังเกตว่าพบว่า ข้อมูล Input x นั้นจะต้องเป็นตัวเลข แต่ในความเป็นจริง เรามักได้ข้อมูลมาเป็นข้อความ เช่น [ใช่, ไม่ใช่] หรือ [น้อย, กลาง, มาก] ถ้าเรามีข้อมูลแบบนี้ เราจะต้องแปลงข้อมูลที่เป็นข้อความหรือหมวดหมู่ ให้เป็นตัวเลขเสียก่อน

สมมุติเรามีข้อมูลรายได้ ระดับความสุข ภูมิภาคที่อาศัย และอายุขัย ทั้งหมด 20 ตัวอย่าง เราต้องการพยากรณ์อายุขัยโดยใช้ข้อมูล 3 อย่างแรก:

Income
Happiness
Area
LifeExpectancy

132546
High
BKK
80

190998
Medium
NE
75

49308
Medium
S
72

32062
Low
NE
59

71707
High
BKK
65

(แสดง 5 ตัวอย่างแรก)

จะเห็นว่า:

  • Income หรือรายได้ เป็นตัวเลขอยู่แล้ว
  • Happiness หรือระดับความสุข เป็นหมวดหมู่ มี 3 ค่า คือ [High, Medium, Low]
  • Area หรือภูมิภาคที่อาศัย เป็นหมวดหมู่ มี 4 ค่า คือ [BKK, NE, S, N]
  • LifeExpectacy เป็นตัวเลขอยู่แล้ว

หน้าที่ของเราคือจะต้องแปลง Happiness และ Area เป็นตัวเลข โดยหลักการมี 2 ขั้น ได้แก่

Factorisation

คือการแทนค่าในแต่ละหมวดด้วยตัวเลขต่างๆ กัน เช่น High = 0, Medium = 1, Low = 2 เป็นต้น โดยเราสามารถแปลงเองใน Excel หรือจะให้ pandas แปลงให้ก็ได้ ลองเขียนโค้ดตั้งแต่เริ่มต้นดังนี้:

import numpy as np
import pandas as pd
from sklearn.preprocessing import OneHotEncoder
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn import metrics
import matplotlib.pyplot as plt

# Load the dataset
df = pd.read_csv("data/life_expectancy.csv")

# Explore
print(df.head())
print(df.info())
print(df.describe())

เราจะเห็นว่า Income และ LifeExpectancy เป็นจำนวนเต็ม (int64) ส่วน Happiness และ Area เป็น object ซึ่งสองตัวนี้เป็นสิ่งที่เราต้องจัดการ

RangeIndex: 20 entries, 0 to 19
Data columns (total 4 columns):
Income            20 non-null int64
Happiness         20 non-null object
Area              20 non-null object
LifeExpectancy    20 non-null int64
dtypes: int64(2), object(2)

จากนั้นให้เลือกเฉพาะคอลัมน์ที่เป็น object แล้วลองดูข้อมูลที่ถูกเลือก

# Limit to categorical data using df.select_dtypes()
df_categories = df.select_dtypes(include=[object])
print(df_categories.head())
print(df_categories.columns)
print(df_categories.shape)

จะได้ Dataframe ใหม่ที่มีคอลัมน์ Happiness และ Area เท่านั้นตามต้องการ และมีมิติ (20, 2)

Index(['Happiness', 'Area'], dtype='object')
(20, 2)

จากนั้นเรียก Class LabelEncoder() ใส่เป็น Argument ของ Method .apply ของ Dataframe ใหม่ของเรา แล้วพ่วง Method .fit_transform เพื่อแปลง scipy sparse matrix ให้เป็น numpy dense matrix ซึ่งอยู่ในรูปแบบที่เราจะใช้

# Create a LabelEncoder object and fit it to each feature in X
X_label = df_categories.apply(LabelEncoder().fit_transform)
print(X_label.head())

ถ้างง บรรทัดนี้คือการสั่งให้ Dataframe ของเราดำเนินการเข้ารหัส Label จากข้อความให้กลายเป็นตัวเลข แล้วแปลงผลลัพธ์ให้อยู่ในรูปแบบที่เอาไปใช้ต่อในการเทรนโมเดลได้

ได้ผลคือ:

   Happiness  Area
0          0     0
1          2     2
2          2     3
3          1     2
4          0     0

ถ้าลองเทียบกลับกับข้อมูลจริง เราจะได้คู่รหัสดังนี้:

  • {0: “Happiness-Hi”, 1: “Happiness-Lo”, 2: “Happiness-Mid”}
  • {0: “Area-BKK”, 1: “Area-N”, 2: “Area-S”, 3: “Area-NE”}

จากนั้นไปสู่ขั้นตอนต่อไป คือการเข้ารหัส One-hot encoding

One-hot encoding

ขั้นตอนที่แล้ว เราได้รหัสตัวเลขเดี่ยวๆ แทน Label ของแต่ละคอลัมน์ จริงๆ ดูเหมือนเราน่าจะนำรหัสนี้ไปใช้เป็น Input ของโมเดลได้เลย ซึ่งจริงในบางกรณี อธิบายคือการเข้ารหัสแบบ Factorisation ตัวเลขที่อยู่ใกล้กันจะถูกตีความว่ามี “คุณค่า” ใกล้กัน เช่น 1 กับ 2 ใกล้กันมากกว่า 1 กับ 3 ดังนั้น ถ้าค่าจริงของ 1, 2, 3 มีความสัมพันธ์เชิงคุณค่าที่ต่อเนื่องกัน เราก็สามารถใช้ Factorisation ได้เลย เช่นถ้า 1 แปลว่า “น้อย”, 2 แปลว่า “ปานกลาง”, และ 3 แปลว่า “มาก” เป็นต้น

แต่ในความเป็นจริง มีหลายกรณีที่ค่าจริงของแต่ละเลขนั้นไม่ได้มีความสัมพันธ์ต่อเนื่องเป็นลำดับขั้นกัน เช่นในกรณีภูมิภาคที่อยู่อาศัย ค่าของแต่ละเลขล้วนมีความหมายของมันเอง และไม่เกี่ยวข้องทางลำดับชั้นกับค่าอื่น

วิธีการแก้ปัญหานี้ คือการใช้ One-hot encoding ซึ่งคือการเข้ารหัสแบบ Binary แทนแต่ละ Label ตัวอย่างเช่น [1, 0, 0, 0] แทน BKK (กทม.), [0, 1, 0, 0] แทน N (ภาคเหนือ) เป็นต้น วิธีการนี้จะทำให้ Label แต่ละอันนั้นเป็นอิสระต่อกัน ทำให้การสร้างโมเดลนั้นแม่นยำมากขึ้น

วิธีการทำ One-hot encoding คือ:

# Create a OneHotEncoder object, and fit it to all of X
X_1hot = OneHotEncoder().fit_transform(X_label).toarray()
print(X_1hot[0:5, :])
print(X_1hot.shape)

ได้ผลคือ:

[[1. 0. 0. 1. 0. 0. 0.]
 [0. 0. 1. 0. 0. 1. 0.]
 [0. 0. 1. 0. 0. 0. 1.]
 [0. 1. 0. 0. 0. 1. 0.]
 [1. 0. 0. 1. 0. 0. 0.]]
(20, 7)

สังเกตว่าการทำ One-hot encoding จะเป็นการขยายจำนวน Feature โดยจำนวนที่ขยาย เท่ากับ ผลรวมของจำนวน Label ที่เป็นไปได้ทั้งหมด ดังนั้นจากตัวอย่างของเรา มี 2 Feature ที่ต้องทำ One-hot encoding, Feature แรกมี 3 Label, Feature ที่สองมี 4 Label ดังนั้นจำนวน Feature ใหม่จึงเท่ากับ 7

คำถามคือ แล้วเราจะเอา Array ใหม่นี้ ไปรวมกับข้อมูลเดิมเพื่อสร้าง Input X ได้อย่างไร คำตอบอบู่ที่ส่วนถัดไป

Re-creating input X

ทบทวนว่าตอนนี้เรามี Dataframe อยู่ 2 ชุด

  • df คือ Dataframe ต้นฉบับ มีหัวข้อคือ Income, Happiness, Area, และ LifeExpectancy
  • df_1hot คือ Dataframe เฉพาะ Happiness และ Area ที่ถูกเข้ารหัส (และขยายขนาด) ด้วย One-hot encoding แล้ว

ทีนี้เรารู้ว่าการจะเทรนโมเดล เราจะต้องเตรียม Dataframe 2 ชุด คือ X ซึ่งเป็น Input ทั้งหมด และ y ซึ่งเป็น Label คำตอบทั้งหมด จะทำอย่างไรดี? อธิบายไปตอนนี้อาจจะงง ลองมาดูโค้ดแล้วอธิบายด้วยโค้ดดีกว่า:

# Recombine numerical and one-hot dataframes to make X
df_1hot = pd.DataFrame(X_1hot)
X = pd.concat([df["Income"], df_1hot], axis=1, sort=False,)
X.columns = ["Income", 
             "Happiness-Hi", "Happiness-Lo", "Happiness-Mid", 
             "Area-BKK", "Area-N", "Area-S", "Area-NE"]
y = df["LifeExpectancy"]
print(X.head())
print(X.shape)
print(y.head())
print(y.shape)

อธิบายดังนี้:

  • เราจะรวม Dataframe เข้าด้วยกัน ดังนั้น ขั้นแรกเราต้องแปลง X_1hot ที่เป็น Array ให้เป็น pandas dataframe ก่อน ตั้งชื่อว่า df_1hot
  • จากนั้นเราประกอบ Dataframe ใหม่ที่จะเป็น X ของเรา ด้วยการ “ต่อ” (Concatenate) คอลัมน์ Income ของ df ต้นฉบับ เข้ากับ df_1hot ทั้งหมด โดยกำหนด Argument axis=1 เพื่อให้ต่อออกไปทางแนวนอน
  • เพื่อความไม่สับสนและเอาไว้อ้างอิง เราจะตั้งชื่อคอลัมน์ใหม่ให้ Dataframe X ของเรา เช่น รหัสแรกของ Happiness ถ้ามีเลข 1 ตรงนั้น ([1, 0, 0]) แปลว่า Happiness-Hi เป็นต้น
  • สุดท้ายก็สร้าง Dataframe y โดยเลือกเฉพาะคอลัมน์ที่เป็นคำตอบ จาก Dataframe df ต้นฉบับ

อย่าลืมตรวจดูว่าผลที่ได้นั้นเป็นไปตามที่เราคิด ด้วยการ print head และ shape ของ X กับ y ออกมาดู

เป็นอันว่าเราได้ Dataframe X และ y ที่พร้อมสำหรับการสร้างโมเดลแล้ว

Model training

เรามาฝึกโมเดล Linear regression กันเลย โดนอย่าลืมแยกข้อมูลออกเป็น Train set กับ Test set ก่อน:

# Split the data into train and test set
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)
print("X_train shape is:", X_train.shape)
print("y_train shape is:", y_train.shape)
print("X_test shape is:", X_test.shape)
print("y_test shape is:", y_test.shape)

# Train the model
lr = LinearRegression().fit(X_train, y_train)
coeff_df = pd.DataFrame(lr.coef_, X.columns, columns=['Coefficient'])
print(coeff_df)

พอฝึกแล้ว มาลองดูว่าโมเดลให้คำตอบเป็นค่า Coefficient ของน้ำหนักสำหรับแต่ละ Feature อย่างไร สำหรับ Linear regression หลายตัวแปร หรือที่เรียกว่า Multivariate linear regression นี้ สูตร Linear function คือ:

โดย n คือจำนวน Feature ซึ่งในกรณีของเราคือ 8

เราสามารถสร้างตารางแสดงค่า Coefficient ของแต่ละ Feature ได้ด้วย pandas dataframe โดยกำหนดใน Argument ให้ lr.coef_ เป็น Input, ให้ชื่อคอลัมน์ X.columns เป็น Label แกนนอน และตั้งชื่อคอลัมน์โดยการกำหนด columns=['Coefficient'] ได้ผลดังนี้:

               Coefficient
Income            0.000055
Happiness-Hi      8.306516
Happiness-Lo     -8.859939
Happiness-Mid     0.553423
Area-BKK         -9.442937
Area-N            3.505052
Area-S            4.850387
Area-NE           1.087498

จากนั้น เราลองมาดูเลยดีกว่าว่าค่าที่พยากรณ์ได้เทียบกับค่าจริงใน Test set เป็นอย่างไร:

# Make a prediction on test set
y_pred = lr.predict(X_test)
df_pred = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})
print(df_pred)

เราพยากรณ์ y_pred โดยใช้ X_test เป็น Input เพื่อให้ y_pred ที่ได้ สามารถนำไปเปรียบเทียบกับ y_test ที่เรารู้ค่าอยู่แล้ว เราป้อนทั้งสองตัวเป็น Input ของ pandas dataframe จะได้ผลว่า:

    Actual  Predicted
3       59  63.213900
16      89  74.945451
6       49  60.294754
10      76  82.234579
2       72  69.815254

ผลที่ได้ไม่ไกล แต่ก็ไม่ถึงกับใกล้ เราอยากรู้เพิ่ม จึงลองคำนวนค่าเฉลี่ยของ y_test และ y_pred ดู:

print("Mean of y_test is", np.mean(np.array(y_test)))
print("Mean of y_pred is", np.mean(y_pred))

ได้ผลว่า:

Mean of y_test is 69.0
Mean of y_pred is 70.10078755113827

ซึ่งใกล้กันมาก แสดงว่าการพยากรณ์แต่ละรายการ มีความแปรผันมาก เลยลองพล็อตกราฟเทียบให้เห็นๆ:

# Plot the conparison between actual and predicted y
df_pred.plot(kind="bar", figsize=(9,7))
plt.show()

ออกมาดังนี้:

Evaluation metrics

การดูด้วยตาย่อมไม่เพียงพอ เราเลยต้องลองใช้มาตรวัดต่างๆ สำหรับ Linear regression มาวัดดู:

# Evaluate the model
print("Train set R2-score = " + str(lr.score(X_train, y_train)))
print("Test set R2-score = " + str(lr.score(X_test, y_test)))

print('Test set Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))
print('Test set Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))
print('Test set Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))

ผลที่ได้มีดังนี้:

Train set R2-score = 0.5401999191488698
Test set R2-score = 0.5965533912121594
Test set Mean Absolute Error: 7.596505622473046
Test set Mean Squared Error: 77.30037024375025
Test set Root Mean Squared Error: 8.792062911726136

อธิบายดังนี้

1) Score วัดค่าสะสมของความต่างระหว่างค่าจริงกับค่าที่พยากรณ์ได้ เมื่อเทียบกับค่าสะสมระหว่างค่าจริงกับค่าเฉลี่ยของค่าจริงทั้งหมด:

2) Mean Absolute Error (MAE) วัดค่าเฉลี่ยของผลรวมความต่างระหว่างค่าจริงกับค่าที่พยากรณ์ได้:

วัดค่าเฉลี่ยของความคลาดเคลื่อนของทั้งโมเดล

3) Mean Squared Error (MSE) วัดค่าเฉลี่ยของผลรวมยกกำลังสองของความต่างระหว่างค่าจริงกับค่าที่พยากรณ์ได้:

สำหรับ MSE ตัวอย่างบางตัวที่มีความคลาดเคลื่อนระหว่างค่าจริงกับค่าพยากรณ์สูง จะถูกให้ค่าน้ำหนักมากเป็นทวีคูณ ดังนั้น MSE จึงชี้วัดได้ดีว่าโมเดลของเรามีตัวอย่างที่มีค่าความคลาดเคลื่อนสูงผิดปกติหรือไม่ (Sensitive to outliers)

4) Root Mean Squared Error (RMSE) คือ Square root ของ MSE:

RMSE ใช้ Square root แปลง MSE กลับมาให้มีหน่วยใกล้เคียงกับ MAE เพื่อให้ตีความเปรียบเทียบได้ง่ายขึ้น สามารถชี้วัดการกระจายตัวของค่าความคลาดเคลื่อนได้ดี ในกรณีของเราให้ค่าใกล้เคียงกับ MAE คือประมาณ ~ 7-8

จะเห็นว่าโมเดลของเรายังพยากรณ์ได้ไม่ดีนัก ซึ่งมีสาเหตุและทางแก้ดังนี้:

  • โมเดลเทรนบนข้อมูลขนาดเล็กมาก คือ Train set 15 รายการเท่านั้น Machine learning ต้องการข้อมูลจำนวนมากกว่านั้นมาก อย่างต่ำๆ ควรจะเป็นหลักหลายๆ ร้อย หรือหลักพัน โดยสำหรับ Linear regression ที่เส้นพยากรณ์เป็นเส้นตรง การที่มีข้อมูลจำนวนมากจะทำให้ค่าความคลาดเคลื่อนโดยเฉลี่ยนั้นลดลง ปัญหานี้เรียกว่าปัญหา Bias ซึ่งตรงข้ามกับปัญหา Variance ที่เป็นโมเดลที่ฟิตกับ Train set มากแต่พยากรณ์ Test set ได้ไม่ดี เราจะพูดถึงปัญหาเหล่านี้ในบทถัดๆ ไป
  • ข้อมูลอาจจะมีความแปรผันมากจนสมการเส้นตรง ไม่สามารถฟิตกับข้อมูลได้ดี ถ้าเรามีข้อมูลมากขึ้น เราอาจะลองพล็อต Scatterplot เพื่อดูรูปร่างความสัมพันธ์ข้อมูล แล้วเปลี่ยนสูตร Algorithm ให้ใช้ Hypothesis function ที่เหมาะสมกับรูปร่างข้อมูลขึ้น เช่น หากพบว่าความสัมพันธ์ใน Scatterplot เป็นเส้นโค้ง อาจพิจารณาใช้ Quadratic function ในรูปแบบ เรียกว่า Polynomial regression ซึ่งสามารถหา Class ใน scikit-learn ใช้ได้

ภาพโดย Skbkekas – Own work, CC BY 3.0

เป็นอันว่าเราสามารถแปลงข้อมูลที่เป็นข้อความหรือหมวดหมู่ให้เป็นตัวเลขได้แล้ว บทต่อไปเราจะพูดถึงการทำ Feature scaling ซึ่งเป็นอีกขั้นตอนของการเตรียมข้อมูล

หน้าแรก | บทที่ 4 Logistic Regression | บทที่ 6 Feature Scaling

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

[NEW] PANTIP.COM : K11950624 มีเรื่องมารบกวนนิดนึงอะค่ะ มีท่อนที่แปลไม่ได้อยู่นิดหน่อยอะค่ะ [ห้องเรียนภาษาอังกฤษ] | shape แปลว่า – NATAVIGUIDES

ความคิดเห็นที่ 9

ถึงแม้ว่าจะเป็นไปได้ที่นักเขียนเก่งๆจะเขียนให้คำศัพท์ตัวใดตัวหนึ่งมีความหมายพลิกแพลงไปได้ (figurative use) แต่ในการตีความต้นฉบับภาษาอังกฤษเพื่อการแปลเป็นไทยนั้น ก่อนที่เราจะมองไปไกลถึงขนาดที่ว่ามันเป็น figurative use หรือเปล่านั้น เราควรสำรวจดูการใช้งานพื้นฐานทั่วๆไปก่อน  ถ้าสำรวจดูดีๆแล้วใช้ precise logic บ่อยครั้งจะพบว่า “grammar เป็นตัวตัดสิน!”

อย่างเช่น

วิธีดูง่ายๆว่า shape แปลว่าอะไร ก็คือดูว่า “object ของ shape เป็นสิ่งที่จับต้องได้หรือเปล่า”

Shape the dough into a ball.
This tool is used for shaping wood.
^
สองประโยคนี้ dough (ลูกแป้ง) และ wood (ไม้) จับต้องได้  shape ก็เลยแปลว่า “จัดรูปทรง”

แต่
His ideas had been shaped by his experiences during the war.
She had a leading role in shaping party policy.
^
สองประโยคนี้ ideas (ความคิด) และ policy (นโยบาย) จับต้องไม่ได้  shape ก็เลยแปลว่า “มีอิทธิพลต่อ”

ประโยคตัวอย่าง 4 ประโยคข้างบนเอามาจาก Oxford Advanced Learner’s Dictionary ตาม link ข้างล่างนี้
http://oald8.oxfordlearnersdictionaries.com/dictionary/shape_2

ในประโยคนี้
Fashion is more than a word…it shapes our everyday way of living… not to mention our behaviour too.
เราจึงตีความว่า
คำว่า “shapes” ต้องแปลว่า “มีอิทธิพลต่อ”

ส่วนคำว่า not to mention  ก็คือ
used to introduce extra information and emphasize what you are saying

He has two big houses in this country, not to mention his villa in France.
คำอธิบายเอามาจาก Oxford Advanced Learner’s Dictionary ตาม link ข้างล่างนี้
http://oald8.oxfordlearnersdictionaries.com/dictionary/mention
^
ในเมื่อมันใช้ “แสดงข้อมูลเพิ่มเติมเพื่อเน้น”

ดังนั้นแทนที่จะแปล “not” เป็น “ไม่” ประโยคนี้เราน่าจะแปล

เปฺ็น
He has two big houses in this country, not to mention his villa in France.
เขามีบ้านสองหลังในประเทศนี้ เท่านั้นไม่พอนะ เขายังมี villa ในฝรั่งเศสอีกด้วย

ลองแปลประโยคของ จขกท ดูดิ
Fashion is more than a word…it shapes our everyday way of living… not to mention our behaviour too.

แฟชั่นเป็นอะไรเกินกว่าที่จะบรรยายด้วยคำพูดได้ แฟชั่นมีอิทธิพลต่อวิถีชีวิตประจำวันของเรา นอกจากนี้แล้วยังมีอิทธิพลต่อพฤติกรรมของเราด้วย

………………………………………………………………
เรื่อง grammar เป็นตัวตัดสินในการตีความนี่เราเคยเจอ classic example ที่ทำให้เรานั่งงงอยู่นานมากๆ (นั่งคิดเป็นชั่วโมง) ตอนแปลนิยายเล่มหนึ่ง นั่นก็คือ verb ที่เราเจอ คือ carry เราหลงคิดว่ามันเป็น transitive verb ได้เท่านั้น เราเลยไล่หา object ในประโยคแล้วเจอ “คำที่ต้องสงสัยตั้ง 3 คำแน่ะว่าจะต้องตีความยังไง”
^
แต่พอไปค้นเจอใน dictionary พบว่า เวลาใช้ sound + carries คำว่า carries กลายเป็น intransitive verb ไป! ส่งผลให้เราต้องกลับไปแก้คำแปลเราแบบรื้อประโยคใหม่หมดเลย

อย่างนี้เขาเรียกว่า “บริบททำให้งง” แต่ “grammar ฟันธงทำให้หายงง”  

จากคุณ
:
fortuneteller

เขียนเมื่อ
:
12 เม.ย. 55 11:07:55


Shapes Song – 31 Kids Songs and Videos | CoComelon Nursery Rhymes \u0026 Kids Songs


With classical and original tunes, this video is a real treat! Babies, toddlers, preschoolers, and kindergartners can learn their shapes in a fun and interactive way. Come sing and dance with us as we explore each shape!
Subscribe for new videos every week: https://www.youtube.com/c/cocomelon?sub_confirmation=1
About Cocomelon:
Where kids can be happy and smart!
At Cocomelon, our goal is to help make learning a fun and enjoyable experience for kids by creating beautiful 3D animation, educational lyrics, and toetapping music.
Kids will laugh, dance, sing, and play along with our videos, learning letters, numbers, animal sounds, colors, and much, much more while simply enjoying our friendly characters and fun stories.
We also make life easier for parents who want to keep their kids happily entertained, giving you the peace of mind that your children are receiving quality educational content. Our videos also give you an opportunity to teach and play with your children as you both watch!
WEBSITE: http://www.Cocomelon.com
FACEBOOK: https://www.facebook.com/Cocomelonkids
INSTAGRAM: https://www.instagram.com/cocomelon_official/
TWITTER: https://www.twitter.com/Cocomelonkids

Copyright © Treasure Studio, Inc. All rights reserved.

นอกจากการดูบทความนี้แล้ว คุณยังสามารถดูข้อมูลที่เป็นประโยชน์อื่นๆ อีกมากมายที่เราให้ไว้ที่นี่: ดูความรู้เพิ่มเติมที่นี่

Shapes Song - 31 Kids Songs and Videos | CoComelon Nursery Rhymes \u0026 Kids Songs

shapes song | shapes rhymes | we are shapes | shape song | shape songs for kids | Kids TV


Hello kids! Do you know what these are? We can see them all around us everyday. We love to make our drawings around these things. Do you know what they are? That’s right you smart kids! They are shapes! Do you know the various shapes which we see around us everyday? Yes, that’s absolutely right kids! We see squares, circles, hearts all around us all the time. But do you know all the shapes which we see everyday? Watch the video above and identify the shapes which we see around us all the time.
This nursery rhyme collection has all the kindergarten favorites. But if we’ve missed anything, it’s only because we’re busy preparing more preschool compilations of the best English rhymes and kid songs we have to offer.
🌈 Hello toddlers, Bob The Train is here to make your learning time easy and fun with these playful toys. Click on the link to explore the toys now! https://amzn.to/2PCeSDS
🌸 Share this video – shapes song here’s the link: https://youtu.be/QBD7CBrroo
🌸 Subscribe for free now to get notified about new videos https://www.youtube.com/kidstv?sub_confirmation=1
🌸 Nursery Rhymes and Nursery Rhymes Karaoke Playlist https://www.youtube.com/watch?v=QNE3PsGVzN8\u0026list=PLDt4VQajKv8xVf8zxjuz3tAdJbCgT2wr
🌸 If you enjoyed this collection, you may also like these compilations:
🌸Old Macdonald had a farm https://www.youtube.com/watch?v=Rf8e4JF8nXs
🌸The Wheels on the bus https://www.youtube.com/watch?v=MWyordkvD0k
🌸Five little monkeys https://www.youtube.com/watch?v=VAL4w6JvDHc
⭐Finger Family https://www.youtube.com/watch?v=IiA17oFD1bs
⭐Ten in the bed https://www.youtube.com/watch?v=Rxxsvc2XXxY
⭐Phonics Song https://www.youtube.com/watch?v=pFR_8zHCLio
🎶 Incy Wincy Spider https://www.youtube.com/watch?v=_6fwkxtp4bw
🌼English Wheels on the bus https://www.youtube.com/watch?v=IEje6DRukxg
🌼Five little ducks https://www.youtube.com/watch?v=vp5rwesz8eU
🌼Johny johny yes papa https://www.youtube.com/watch?v=PyZq_Jm0aqQ
🌼 Humpty Dumpty https://www.youtube.com/watch?v=ZSoougnleU8
🌼 Twinkle Twinkle Little Star https://www.youtube.com/watch?v=yvcrrmeBwlo
🌼 Ten in the bed https://www.youtube.com/watch?v=GRyxRadTM3c
Nursery rhymes and kid’s songs accelerate phonetic awareness improving children’s word comprehension, reading and writing skills. These rhymes for children help teach basic skills and improves their ability to comprehend and follow directions.
We hope you’re having a fun time with all your friends here at Kids TV. If you enjoyed watching this video then check out our channel for many more interesting and fun learning videos for kids.
If you’re still reading this far, we know you enjoy our animations but are always happy to hear from you on how we can improve and what you’d like to see in the pipeline!
If you enjoy our content, don’t forget to support us and subscribe 🙂
Download the Bob the Train APP here:
Bob The Train: https://itunes.apple.com/us/app/bobthetrainkidsnurseryrhymesbabysongs/id1236462400?ls=1\u0026mt=8
Kids First: https://itunes.apple.com/us/app/kidsfirstkidsvideosnurseryrhymes/id1245088728?ls=1\u0026mt=8
FB: https://www.facebook.com/KidsTVNurseyRhymes
Web: https://www.uspstudios.co/creation/channel/kidstv/1/
Twitter: https://twitter.com/kidstvofficial
G+: https://plus.google.com/u/0/+kidstv
Instagram: https://www.instagram.com/kidstvofficial/
Pinterest: https://in.pinterest.com/kidstv06/pins/
Play store:
Bob The Train: https://play.google.com/store/apps/details?id=com.uspstudios.bobthetrain\u0026pageId=108868050184468649173
Kids First: https://play.google.com/store/apps/details?id=com.lookedigital.kidsfirst
Subscribe: https://www.youtube.com/channel/UC7Pq3Ko42YpkCB_Q4E981jw?sub_confirmation=1
Click on the Link Below to Watch and Read Wonderful Stories from Kids TV only on Rivet! :
https://rivet.area120.com/read//category/youtube/Kids%20TV?utm_source=youtube_direct\u0026utm_campaign=Kids%20TV
© 2017 USP Studios Private Limited
============================================
Music and Lyrics: Copyright USP Studios™
Video: Copyright USP Studios™
============================================

shapes song | shapes rhymes | we are shapes | shape song | shape songs for kids | Kids TV

School Supplies Song


It’s a fun School Supplies Song for kids! Teach and learn English classroom words with this original music video. FREE WORKSHEET included! Great for young children, ESL/EFL students, or babies. It’s EASY to sing along and young learners will love it. Learn the words pencil, sharpener, eraser, ruler, book, scissors, chair, desk, pen, and bag.
Get a FREE WORKSHEET just for this video here at our website:
http://www.englishtreetv.com/schoolsuppliessongresources.html

Music and animation written, performed, and made by Adam WilliamsWalters.
Copyright 2015 Adam WilliamsWalters/English Tree TV, All rights reserved.

Other Places To Watch:
► AMAZON VIDEO
Are you an Amazon Prime member? Watch English Tree TV videos adfree on Amazon Prime Video. Just search for “English Tree TV”
► KIDOODLE.TV https://kidoodle.tv/
Other Places to Listen:
► SPOTIFY https://spoti.fi/3hdZuLF
► iTUNES https://apple.co/2XRZ7Pn
FREE TEACHING RESOURCES: http://www.englishtreetv.com
SOCIAL MEDIA:
Facebook: https://www.facebook.com/englishtreetv
Instagram: https://www.instagram.com/englishtreetv/
Subscribe to English Tree TV!
http://www.youtube.com/channel/UC4GaQ9fCH5IpAq1AImj7XuQ?sub_confirmation=1

School Supplies Song

คำศัพท์ภาษาอังกฤษ รูปทรงเรขาคณิต (Geometric shapes)


++ คำศัพท์ภาษาอังกฤษ รูปทรงเรขาคณิต (Geometric shapes) ++
รูปทรงเรขาคณิตภาษาอังกฤษ
1. Circle (เซอเคิล) รูปวงกลม
2. Oval (โอเวิล) รูปวงรี
3. Triangle (ไทรแองเกิล) รูปสามเหลี่ยม
4. Square (สเกวเออะ) รูปสี่เหลี่ยมจัตุรัส
5. Rectangle (เรคแทงเกิล) รูปสี่เหลี่ยมผืนผ้า
6. Pentagon (เพนทากอน) รูปห้าเหลี่ยม
7. Hexagon (เฮ็คซากอน) รูปหกเหลี่ยม
8. Octagon (ออคทากอน) รูปแปดเหลี่ยม
9. Arrow (แอโร) ลูกศร
10. Crescent (เครสเซนท์) รูปจันทร์เสี้ยว
11. Cross (ครอซ) รูปกากบาท
12. Diamond (ไดมอนด์) รูปสี่เหลี่ยมขนมเปียกปูน
13. Heart shape (ฮาทเชฟ) รูปหัวใจ
14. Star shape (สทาเชฟ) รูปดาว
15. Semicircle (เซมิเซอเคิล) รูปครึ่งวงกลม
++ จัดทำโดย Por Pure Channel ++
ฝากกด Like กดแชร์ คลิปวีดีโอ และกด Subscribe ด้วยจร้า
จะอัพคลิปเรื่อยๆ
Channel : https://www.youtube.com/channel/UCuxFOCWjUZ7PxmzBaxialBQ
Subscribe : https://www.youtube.com/channel/UCuxFOCWjUZ7PxmzBaxialBQ?sub_confirmation=1
คำศัพท์อังกฤษ ภาษาอังกฤษสำหรับเด็กน้อย คำศัพท์อังกฤษง่ายๆ

คำศัพท์ภาษาอังกฤษ รูปทรงเรขาคณิต (Geometric shapes)

Shapes l รูปร่าง รูปทรง | เรียนคำศัพท์ภาษาอังกฤษฟรี


คลิปนี้มาสั้นๆ แต่โดนใจแน่นอนค่ะ โดยเฉพาะสายคณิตศาสตร์ต้องรู้แล้วล่ะค่ะว่าคำศัพท์เกี่ยวกับรูปร่าง/รูปทรงในภาษาอังกฤษเรียกว่ายังไงบ้าง
ขอบอกเลยว่าทำคลิปนี้ไป ทัยก็ได้เรียนรู้คำศัพท์ใหม่ๆไปด้วย ว้าวมากๆเลยค่ะ
ไปดูกันเลยยยย

Shapes l รูปร่าง รูปทรง | เรียนคำศัพท์ภาษาอังกฤษฟรี

นอกจากการดูบทความนี้แล้ว คุณยังสามารถดูข้อมูลที่เป็นประโยชน์อื่นๆ อีกมากมายที่เราให้ไว้ที่นี่: ดูบทความเพิ่มเติมในหมวดหมู่LEARN FOREIGN LANGUAGE

ขอบคุณที่รับชมกระทู้ครับ shape แปลว่า

Leave a Reply

Your email address will not be published. Required fields are marked *