AWS+darknet install using python code.
Budget: $250 – $750 USD
We are looking the python program specialist for learning darknet on AWS .
using Python boto3.
The requirements are as follows.
1. Generate a GPU Instance (called EC2GPU) on AWS with a command, set training parameters on the local GUI, and send training data and parameters to AWS.
Create EC2 Instabce using CLI.
EC2 can select GPU, number of CPUs, number of memories, HDD / SDD capacity.
For parameters such as GPU type, get a list for each AWS access .
2. Generate EC2 with the set parameters and start learning
During learning on AWS, it will be executed even if there is no connection, and only the status will be displayed on the local GUI.
Being able to check the learning progress, CPU time, elapsed time, end time, Loss, mAP, etc.
3. When learning is completed, AWS can notify the end of learning status.
SNS,SMTP,SQS ok
4. After the learning is completed, transfer the generated learning data from AWS to the local GUI and EC2 GPU instance remove or stop.
5. Display the expected pay-as-you-go amount and the current amount (calculated OK)
6. Delete EC2 instance on AWS.
The requirements specifications are summarized as follows.
1. Start learning on the local PC
2. instance create -> darknet install
3 .after darknet install
4. localpc-> train files upload EC2
5. after upload -> AWS train start
6. if train finished-> local pc notify
7. train result file download -> local pc
8. AWS instance remove.
using Python boto3.
The requirements are as follows.
1. Generate a GPU Instance (called EC2GPU) on AWS with a command, set training parameters on the local GUI, and send training data and parameters to AWS.
Create EC2 Instabce using CLI.
EC2 can select GPU, number of CPUs, number of memories, HDD / SDD capacity.
For parameters such as GPU type, get a list for each AWS access .
2. Generate EC2 with the set parameters and start learning
During learning on AWS, it will be executed even if there is no connection, and only the status will be displayed on the local GUI.
Being able to check the learning progress, CPU time, elapsed time, end time, Loss, mAP, etc.
3. When learning is completed, AWS can notify the end of learning status.
SNS,SMTP,SQS ok
4. After the learning is completed, transfer the generated learning data from AWS to the local GUI and EC2 GPU instance remove or stop.
5. Display the expected pay-as-you-go amount and the current amount (calculated OK)
6. Delete EC2 instance on AWS.
The requirements specifications are summarized as follows.
1. Start learning on the local PC
2. instance create -> darknet install
3 .after darknet install
4. localpc-> train files upload EC2
5. after upload -> AWS train start
6. if train finished-> local pc notify
7. train result file download -> local pc
8. AWS instance remove.