homayra22 Goto Github PK

cassava_leaf_detection

Cassava is a significant food security crop grown by smallholder farmers because it can tolerate harsh conditions. It is Africa’s second-largest producer of carbs. This starchy root is grown on at least 80% of home farms in Sub-Saharan Africa, although viral infections are a primary cause of low yields. It may be feasible to identify common diseases and cure them with the help of data science.Image classification methods based on convolutional neural networks have been grown rapidly given the growth of deep learning architecture and hardware computing capacity. In this study an empirical comparison is done to analyze the different model’s performance in terms of image classification. To identify cassava leaf diseases, both classic machine learning methods and deep learning models are used. The results shows that the overall accuracy of the model obtained is about 72% .Cassava is a significant food security crop grown by smallholder farmers because it can tolerate harsh conditions. It is Africa’s second-largest producer of carbs. This starchy root is grown on at least 80% of home farms in Sub-Saharan Africa, although viral infections are a primary cause of low yields. It may be feasible to identify common diseases and cure them with the help of data science.Image classification methods based on convolutional neural networks have been grown rapidly given the growth of deep learning architecture and hardware computing capacity. In this study an empirical comparison is done to analyze the different model’s performance in terms of image classification. To identify cassava leaf diseases, both classic machine learning methods and deep learning models are used. The results shows that the overall accuracy of the model obtained is about 72% .

green_code

This repositories talks about improving Python Code Efficiency for Energy Saving

icebreaker

image-processing

This repository works about some Image Processing Techniques

power-measurement-of-nlp-methods-using-zero-shot-classification

Humans can learn new concepts of tasks in efficient amount of time. With the advancement of technology, it is needed to implement NLP models with different domains of work on huge amount of data. These models have gained noteworthy accuracy in domain based tasks. However, the accuracy of these models depends on the available hardware and software resources on model training and how the training process impacts on environment in terms of power consumption. As a result, these models are costly to develop in particular amount of work. Zero shot classification is the unsupervised method where the model can be implemented on unseen label in multipurpose domains of work. In this study, this issue is brought under attention to select a model with less power consumption with high accuracy implemented on zero shot classification.Humans can learn new concepts of tasks in efficient amount of time. With the advancement of technology, it is needed to implement NLP models with different domains of work on huge amount of data. These models have gained noteworthy accuracy in domain based tasks. However, the accuracy of these models depends on the available hardware and software resources on model training and how the training process impacts on environment in terms of power consumption. As a result, these models are costly to develop in particular amount of work. Zero shot classification is the unsupervised method where the model can be implemented on unseen label in multipurpose domains of work. In this study, this issue is brought under attention to select a model with less power consumption with high accuracy implemented on zero shot classification.

regression-analysis

Some Regression Analysis Techniques using R language

thesis-project

Road Accident Debris Detection Using Object Detection API