================= Our experiments are conducted using Pytorch 1.4.0====================
To run our experiment over all advanced models, please first activate your pytorch evirenment, and just use:
(torch)$ python3 train_seeds.py
Then you will get accuracy records in model files.
Next, use:
(torch)$ python3 compute_mu_std.py
to do the results processing work.
After this, you will see all_model.json on current fold.
Finally, use:
(torch)$ python3 draw_all.py
to do the visualization work and get mean accuracy and standard deviation.
(torch)$ python3 t-SNE.py
to draw the pictures of case study.
If you want to run our baselines, just open the .ipynb files on fold \baselines.
comprehensiveMap/EI328-project
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|