This is a fixed price project of $40.
We are looking for an expert in CNN and metaheuristics, to write a jupyter python script for automaticlly designing CNN architecture by different swarm-intelligent algorithms, such as ABC, Bat algorithm, Firefly algorithm, whale optimization algorithm on MNIST and CIFAR. Each line should be commented and explained in the script. The script should also contain accuracy vizualization for each experiment, confusion matrix, algorthm comparison, other visualizatoin etc. The results should be stored and saved in csv file. The parameters that should be optimize are the following: number of convolutional layer, number of filters for each convolutional layer, filter size for each convolutional layer, activation function, pooling layer size, number of fully connected layers, number of neurons in each layer, weight regularization in each fc layer, droput rate, batch size, learning rule and learning rate.
When you apply, please specify the time you would need to complete the project and which other visualization technique you would use and which similar project did you do before.
5 фрилансеров(-а) готовы выполнить эту работу в среднем за $226
Hello! I'd like to deliver neural architecture search study in 2 days. I'm familiar with NAS, RL, deep learning, pytorch. I'll do the job blazingly fast. Please, give me a try!