Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 27

Experimentation And Validation Of Results:

Detailed Models and There parameters with sample Results :

1.Crop yield prediction Models :


1.1 random forest regressor
MSE : 684938.2423372074
MAE : 43.61852710509038
R2 Score /Accuracy : -2.89

1.2 SVR
MSE: 5768.43679
MAE : 0.6864786957332968
R2 score : 0.8601401414634322

1.3 Xgboost
MSE: 432256.76543
MAE : 23.7941745977
R2 score : 0.7788534426283292

1.4 Linear Regression


MSE: 720830.0582290915
MAE : 98.51837221
R2 score : 0.14068505458916214

2.Crop Recommendation Models parameters :


2.1.Decision Tree :

Accuracy : 82.72727272727273

Classification report as fallows :


Cross validation Score :

array([0.82727273, 0.82954545, 0.83181818, 0.84090909, 0.83409091])

2.2 Naive Bayes Theorem :

Accuracy : 0.9522727272727273

Classification Report :
Cross Validation accuracy :

array([0.96818182, 0.95 , 0.95909091, 0.93636364, 0.95227273])

2.3 SVM

Accuracy : 0.9318181818181818

Classification Report :
Cross Validation Accuracy :

array([0.78409091, 0.78863636, 0.79545455, 0.75454545, 0.76136364])

2.4 Logistic Regression

Accuracy : 0.6545454545454545

Classification Report :
Cross Validation Parameters :

array([0.66363636, 0.63863636, 0.6 , 0.67272727, 0.66136364])

2.5 Random Forest :

Accuracy : 0.9727272727272728

Classification Report :
Cross validation accuracy : array([0.96363636, 0.96818182, 0.96136364, 0.95454545, 0.95454545])

2.6 Xgboost

Accuracy : 0.9545454545454546

Classification Report :
Cross Validation Accuracy : array([0.95681818, 0.95227273, 0.96363636, 0.94318182,
0.95454545])

Comparative Grap[h :
3.Fertiliser Recommendation :
3.1 Decision tree :

Accuracy : 85.0

Classification report :

Cross validation Accuracy : array([0.95, 1. , 0.9 , 0.95, 1. ])

3.2 Guassian Naive Bayes :


Accuracy : 0.65
Classification report :

Cross validation accuracy : array([0.6 , 0.55, 0.9 , 0.45, 1. ])

3.3 SVM

Accuracy : 0.5
Classification Report :

Cross Validation Accuracy : array([0.95 , 0.75 , 0.8 , 0.8 , 0.89473684])

3.4 Logistic Regression :


Accuracy : 1.0

Classification report :

Cross validation accuracy : array([1., 1., 1., 1., 1.])

3.5 Random Forest

Accuracy : RF's Accuracy is: 0.9

Classification report :

Cross validation accuracy : array([0.85 , 0.95 , 0.95 , 0.9 , 0.84210526])


3.6 XGBosst

Accuracy : 1.0

Classification report :

Cross validation accuracy : array([0.95, 0.95, 1. , 0.95, 1. ])

Comparison graph :
4. Plant Disease Detection :

After traaining for 6 epochs

RESNET50 Architecture.
Accuracy changes graph

Loss changes graph

Test Loss: 0.21739


Test Accuracy: 92.83%
Complete Classification Report.

Experimentation Results :

1.Plant Disease Detection :

Dataset image distribution


The above are few image sample display from the given dataset.
Result predicted and tested on few test sample images.

2.crop Yield :
This above thing is correlation between input column to detect best features.
Graph s howing season names with production.
Graph showing the impact of the pressure on the production.
Impact of humidity on production
Impact of temperature on production.
Impact of area on production .

3.Crop Recommendation :
Nitrogen data change in given dataset

Phosphorus data change in given dataset.


Photosium data in given dataset.

Correlation to detect best features as input.

4.Fertiliser Detection :
The above things show the impact on type of fertiliser we select.

You might also like