You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
242 lines
10 KiB
242 lines
10 KiB
# Vainisha za vyakula 2
|
|
|
|
Katika somo hili la pili la uainishaji, utachunguza njia zaidi za kuainisha data ya nambari. Pia utajifunza kuhusu athari za kuchagua vainisha mmoja badala ya mwingine.
|
|
|
|
## [Jaribio kabla ya mihadhara](https://ff-quizzes.netlify.app/en/ml/)
|
|
|
|
### Sharti
|
|
|
|
Tunadhani umefanya masomo ya awali na una seti safi ya data katika folda yako ya `data` inayoitwa _cleaned_cuisines.csv_ kwa mzizi wa folda hii ya masomo 4.
|
|
|
|
### Maandalizi
|
|
|
|
Tumepakia faili lako la _notebook.ipynb_ lenye seti safi ya data na tumeigawanya kuwa dataframes za X na y, tayari kwa mchakato wa ujenzi wa modeli.
|
|
|
|
## Ramani ya uainishaji
|
|
|
|
Hapo awali, ulijifunza kuhusu chaguzi mbalimbali ulizo nazo wakati wa kuainisha data kwa kutumia karatasi ya hila ya Microsoft. Scikit-learn inatoa karatasi ya hila inayofanana, lakini yenye maelezo zaidi ambayo inaweza kusaidia zaidi kupunguza vainisha wako (neno jingine la vainisha ni makadirio):
|
|
|
|
|
|
> Vidokezo: [tembelea ramani hii mtandaoni](https://scikit-learn.org/stable/tutorial/machine_learning_map/) na bonyeza mfululizo wa njia kusoma nyaraka.
|
|
|
|
### Mpango
|
|
|
|
Ramani hii ni msaada mkubwa mara unapokuwa na uelewa wazi wa data yako, kwani unaweza 'kutembea' kwenye njia zake kuelekea uamuzi:
|
|
|
|
- Tuna sampuli >50
|
|
- Tunataka kutabiri kategoria
|
|
- Tuna data zilizo na lebo
|
|
- Tuna sampuli chini ya 100K
|
|
- ✨ Tunaweza kuchagua Linear SVC
|
|
- Ikiwa hiyo haitumiki, kwa kuwa tuna data za nambari
|
|
- Tunaweza kujaribu ✨ KNeighbors Classifier
|
|
- Ikiwa hiyo haitumiki, jaribu ✨ SVC na ✨ Ensemble Classifiers
|
|
|
|
Hii ni njia yenye msaada mkubwa kufuata.
|
|
|
|
## Zoha - gawanya data
|
|
|
|
Kwa kufuata njia hii, tunapaswa kuanza kwa kuingiza maktaba kadhaa kutumia.
|
|
|
|
1. Ingiza maktaba zinazohitajika:
|
|
|
|
```python
|
|
from sklearn.neighbors import KNeighborsClassifier
|
|
from sklearn.linear_model import LogisticRegression
|
|
from sklearn.svm import SVC
|
|
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
|
|
from sklearn.model_selection import train_test_split, cross_val_score
|
|
from sklearn.metrics import accuracy_score,precision_score,confusion_matrix,classification_report, precision_recall_curve
|
|
import numpy as np
|
|
```
|
|
|
|
1. Gawanya data yako ya mafunzo na mtihani:
|
|
|
|
```python
|
|
X_train, X_test, y_train, y_test = train_test_split(cuisines_features_df, cuisines_label_df, test_size=0.3)
|
|
```
|
|
|
|
## Vainisha la Linear SVC
|
|
|
|
Support-Vector clustering (SVC) ni sehemu ya familia ya Mashine za Support-Vector za mbinu za ML (jifunza zaidi kuhusu hizi hapa chini). Katika njia hii, unaweza kuchagua 'kernel' kuamua jinsi ya kuunganisha lebo. Parameter ya 'C' inahusu 'regularization' ambayo hudhibiti ushawishi wa vigezo. Kernel inaweza kuwa moja ya [nguvu kadhaa](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html#sklearn.svm.SVC); hapa tumeiseti kuwa 'linear' kuhakikisha tunatumia linear SVC. Probability kwa default ni 'false'; hapa tumeiseti kuwa 'true' kukusanya makadirio ya uwezekano. Tumeweka hali ya bahati nasibu kuwa '0' kuchanganya data kupata uwezekano.
|
|
|
|
### Zoha - tumia linear SVC
|
|
|
|
Anza kwa kuunda safu ya vainisha. Utaiongeza polepole kwenye safu hii tunapojaribu.
|
|
|
|
1. Anza na Linear SVC:
|
|
|
|
```python
|
|
C = 10
|
|
# Unda waainishaji tofauti.
|
|
classifiers = {
|
|
'Linear SVC': SVC(kernel='linear', C=C, probability=True,random_state=0)
|
|
}
|
|
```
|
|
|
|
2. Funza modeli yako kwa kutumia Linear SVC na chapisha ripoti:
|
|
|
|
```python
|
|
n_classifiers = len(classifiers)
|
|
|
|
for index, (name, classifier) in enumerate(classifiers.items()):
|
|
classifier.fit(X_train, np.ravel(y_train))
|
|
|
|
y_pred = classifier.predict(X_test)
|
|
accuracy = accuracy_score(y_test, y_pred)
|
|
print("Accuracy (train) for %s: %0.1f%% " % (name, accuracy * 100))
|
|
print(classification_report(y_test,y_pred))
|
|
```
|
|
|
|
Matokeo ni mazuri kweli:
|
|
|
|
```output
|
|
Accuracy (train) for Linear SVC: 78.6%
|
|
precision recall f1-score support
|
|
|
|
chinese 0.71 0.67 0.69 242
|
|
indian 0.88 0.86 0.87 234
|
|
japanese 0.79 0.74 0.76 254
|
|
korean 0.85 0.81 0.83 242
|
|
thai 0.71 0.86 0.78 227
|
|
|
|
accuracy 0.79 1199
|
|
macro avg 0.79 0.79 0.79 1199
|
|
weighted avg 0.79 0.79 0.79 1199
|
|
```
|
|
|
|
## Vainisha wa K-Neighbors
|
|
|
|
K-Neighbors ni sehemu ya familia ya mbinu za ML za "neighbors", ambazo zinaweza kutumika kwa kujifunza kwa uangalizi au bila uangalizi. Katika njia hii, idadi iliyowekwa ya pointi huundwa na data hukusanywa karibu na pointi hizi ili lebo jumla zitabiriwe kwa data.
|
|
|
|
### Zoha - tumia vainisha wa K-Neighbors
|
|
|
|
Vainisha wa awali ulikuwa mzuri, na ulikuwa na utendaji mzuri kwenye data, lakini labda tunaweza kupata usahihi bora zaidi. Jaribu vainisha wa K-Neighbors.
|
|
|
|
1. Ongeza mstari kwenye safu yako ya vainisha (ongeza koma baada ya kipengele cha Linear SVC):
|
|
|
|
```python
|
|
'KNN classifier': KNeighborsClassifier(C),
|
|
```
|
|
|
|
Matokeo ni mabaya kidogo:
|
|
|
|
```output
|
|
Accuracy (train) for KNN classifier: 73.8%
|
|
precision recall f1-score support
|
|
|
|
chinese 0.64 0.67 0.66 242
|
|
indian 0.86 0.78 0.82 234
|
|
japanese 0.66 0.83 0.74 254
|
|
korean 0.94 0.58 0.72 242
|
|
thai 0.71 0.82 0.76 227
|
|
|
|
accuracy 0.74 1199
|
|
macro avg 0.76 0.74 0.74 1199
|
|
weighted avg 0.76 0.74 0.74 1199
|
|
```
|
|
|
|
✅ Jifunze kuhusu [K-Neighbors](https://scikit-learn.org/stable/modules/neighbors.html#neighbors)
|
|
|
|
## Vainisha wa Support Vector
|
|
|
|
Support-Vector vainisha ni sehemu ya familia ya mbinu za ML za [Support-Vector Machine](https://wikipedia.org/wiki/Support-vector_machine) zinazotumika kwa kazi za uainishaji na regression. SVMs "huweka mifano ya mafunzo kwenye pointi katika nafasi" ili kuongeza umbali kati ya vikundi viwili. Data inayofuata huwekwa kwenye nafasi hii ili kategoria yake itabiriwe.
|
|
|
|
### Zoha - tumia Vainisha wa Support Vector
|
|
|
|
Tujaribu usahihi kidogo bora na Vainisha wa Support Vector.
|
|
|
|
1. Ongeza koma baada ya kipengele cha K-Neighbors, kisha ongeza mstari huu:
|
|
|
|
```python
|
|
'SVC': SVC(),
|
|
```
|
|
|
|
Matokeo ni mazuri sana!
|
|
|
|
```output
|
|
Accuracy (train) for SVC: 83.2%
|
|
precision recall f1-score support
|
|
|
|
chinese 0.79 0.74 0.76 242
|
|
indian 0.88 0.90 0.89 234
|
|
japanese 0.87 0.81 0.84 254
|
|
korean 0.91 0.82 0.86 242
|
|
thai 0.74 0.90 0.81 227
|
|
|
|
accuracy 0.83 1199
|
|
macro avg 0.84 0.83 0.83 1199
|
|
weighted avg 0.84 0.83 0.83 1199
|
|
```
|
|
|
|
✅ Jifunze kuhusu [Support-Vectors](https://scikit-learn.org/stable/modules/svm.html#svm)
|
|
|
|
## Vainisha za Ensemble
|
|
|
|
Tufuate njia hadi mwisho kabisa, ingawa jaribio la awali lilikuwa zuri sana. Tujaribu 'Vainisha za Ensemble', hasa Random Forest na AdaBoost:
|
|
|
|
```python
|
|
'RFST': RandomForestClassifier(n_estimators=100),
|
|
'ADA': AdaBoostClassifier(n_estimators=100)
|
|
```
|
|
|
|
Matokeo ni mazuri sana, hasa kwa Random Forest:
|
|
|
|
```output
|
|
Accuracy (train) for RFST: 84.5%
|
|
precision recall f1-score support
|
|
|
|
chinese 0.80 0.77 0.78 242
|
|
indian 0.89 0.92 0.90 234
|
|
japanese 0.86 0.84 0.85 254
|
|
korean 0.88 0.83 0.85 242
|
|
thai 0.80 0.87 0.83 227
|
|
|
|
accuracy 0.84 1199
|
|
macro avg 0.85 0.85 0.84 1199
|
|
weighted avg 0.85 0.84 0.84 1199
|
|
|
|
Accuracy (train) for ADA: 72.4%
|
|
precision recall f1-score support
|
|
|
|
chinese 0.64 0.49 0.56 242
|
|
indian 0.91 0.83 0.87 234
|
|
japanese 0.68 0.69 0.69 254
|
|
korean 0.73 0.79 0.76 242
|
|
thai 0.67 0.83 0.74 227
|
|
|
|
accuracy 0.72 1199
|
|
macro avg 0.73 0.73 0.72 1199
|
|
weighted avg 0.73 0.72 0.72 1199
|
|
```
|
|
|
|
✅ Jifunze kuhusu [Vainisha za Ensemble](https://scikit-learn.org/stable/modules/ensemble.html)
|
|
|
|
Njia hii ya Kujifunza Mashine "inaunganisha utabiri wa makadirio mengi ya msingi" ili kuboresha ubora wa modeli. Katika mfano wetu, tulitumia Miti ya Bahati nasibu na AdaBoost.
|
|
|
|
- [Random Forest](https://scikit-learn.org/stable/modules/ensemble.html#forest), njia ya kusawazisha, hujenga 'msitu' wa 'miti ya uamuzi' umetawanywa kwa bahati nasibu ili kuepuka kufikia kiwango cha juu sana cha kufaa data. Parameter ya n_estimators imewekwa kwa idadi ya miti.
|
|
|
|
- [AdaBoost](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.AdaBoostClassifier.html) huendana na vainisha kwa seti ya data kisha huiga vainisha huo kwa seti ile ile ya data. Inazingatia uzito wa vitu vilivyokosewa na kurekebisha utendakazi kwa vainisha inayofuata kurekebisha kosa.
|
|
|
|
---
|
|
|
|
## 🚀Changamoto
|
|
|
|
Kila moja ya mbinu hizi ina vigezo vingi unaweza kubadilisha. Fanya utafiti wa vigezo vyao vya default na fikiria maana ya kubadilisha vigezo hivi kwa ubora wa modeli.
|
|
|
|
## [Jaribio baada ya mihadhara](https://ff-quizzes.netlify.app/en/ml/)
|
|
|
|
## Mapitio & Kusoma Kibinafsi
|
|
|
|
Kuna maneno mengi magumu katika masomo haya, hivyo chukua dakika moja kupitia [orodha hii](https://docs.microsoft.com/dotnet/machine-learning/resources/glossary?WT.mc_id=academic-77952-leestott) ya istilahi muhimu!
|
|
|
|
## Kazi ya nyumbani
|
|
|
|
[Cheza na vigezo](assignment.md)
|
|
|
|
---
|
|
|
|
<!-- CO-OP TRANSLATOR DISCLAIMER START -->
|
|
**Kiarifa cha Hapana Dhima**:
|
|
Nyaraka hii imetafsiriwa kwa kutumia huduma ya tafsiri ya AI [Co-op Translator](https://github.com/Azure/co-op-translator). Ingawa tunajitahidi kuwa sahihi, tafadhali fahamu kuwa tafsiri za moja kwa moja zinaweza kuwa na makosa au kasoro. Nyaraka asilia katika lugha yake ya asili inapaswa kuzingatiwa kama chanzo cha uhakika. Kwa taarifa muhimu, tafsiri ya kitaalamu ya mwanadamu inashauriwa. Hatuwajibiki kwa kutoelewana au tafsiri za makosa zinazotokana na matumizi ya tafsiri hii.
|
|
<!-- CO-OP TRANSLATOR DISCLAIMER END --> |