mariegold / deeplearning

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Deep Learning Mini-Projects (May 2021)

Authors: Marie Biolkova, Minh Tri Pham

Project 1

Usage

To see the training and accuracy of the best model overall (1 run):

python3 Proj1/test.py 

To see the tuning and accuracies (with standard deviations) of all tested models (10 runs), modify test.py:

# test.py 
run_full = True

and run the same bash command as above. This takes a while to complete (~40 minutes)

Project 2

Usage

To see the training and accuracy of the best model trained by SGD and MSE (1 run):

python3 Proj2/test.py 

To see the training and accuracy of the best model overall (1 run), modify test.py:

# test.py 
run_best = True

and run the same bash command as above.

To see the tuning and accuracies (with standard deviations) of all tested models (10 runs), set:

# test.py 
run_training = True

and run the same bash command as above. This takes a while to complete (~10 minutes).

About


Languages

Language:Python 100.0%