Basically I am having Acer Nitro 50 Desktop with system configuration
Processor: Intel® Core™ i5-8400 CPU @ 2.80GHz × 6
Graphics: GeForce GTX 1050/PCIe/SSE2 (2 GB)
Memory(RAM): 8GB DDR4 Memory
I am working with tensorflow to train a faster_rcnn_inception_v2 model on my custom dataset with 10.4 GB(65000 images) of training data and 533.4 MB(3333) of validation data for object detection for 600k epochs(num_steps it will be training at 640 x 640 resolution). I am training and validation model on 8 classes. So when I am training the model the accuracy(map) is not increasing and loss is not decreasing after a while. After the completion of the training successfully I ran the model on various images and noticed couple of things.
- Less accuracy
- In some images objects does not get detected at all no bounding boxes. while in some multiclass object detection is spot on perfectly all objects gets detected but with less accuracy around 60 - 75.
- From second point above If I train a separate model with less images and less number of classes(3 or 4) it works well but with decent amount of accuracy around 75 - 95
Question 1) What would be ideal system to work comfortably with deep learning? and get desired amount of results
So based on my system configuration should I keep on training or invest into new system which will be lot faster and accurate in deep learning. It would be really helpful for me to continue my learning if anyone could suggest or recommend something