OSPP: Implementation of a Class Incremental Learning Algorithm Evaluation System based on Ianvs #85
Conversation
qxygxt
force-pushed
the
main
branch
2 times, most recently
from
507dfce
to
b7039df
Compare
|
lgtm |
| @@ -0,0 +1,38 @@ | |||
| for i in range(len(nam_label)): | |||
There was a problem hiding this comment.
If this is not a useful file, please remove it in this commit.
| 'best_pred': self.trainer.best_pred, | ||
| }, is_best) | ||
|
|
||
| # if not self.trainer.args.no_val and \ |
There was a problem hiding this comment.
If this is unuseful debug lines, it should not be committed.
| img /= self.std | ||
|
|
||
| # mean and std for original depth images | ||
| mean_depth = 0.12176 |
There was a problem hiding this comment.
what is these magic number meaning? It is suggested to add comment for these number.
| depth = sample['depth'] | ||
| mask = sample['label'] | ||
| width, height = img.size | ||
| left = 140 |
There was a problem hiding this comment.
what is these magic number meaning? It is suggested to add comment for these number.
If these numbers are not the only value of varibles, please config these number in a seperated CONSTANT file.
| img = img.resize((width,height), Image.BILINEAR) | ||
| depth = depth.resize((width,height), Image.BILINEAR) | ||
| mask = mask.resize((width,height), Image.NEAREST) | ||
| # img = img.resize((512,1024), Image.BILINEAR) |
There was a problem hiding this comment.
Suggest to remove these debug lines, Also the following.
| @@ -0,0 +1,50 @@ | |||
| import os | |||
| os.environ['BACKEND_TYPE'] = 'PYTORCH' | |||
| import os | ||
|
|
||
| os.environ['BACKEND_TYPE'] = 'PYTORCH' | ||
| # os.environ["UNSEEN_SAVE_URL"] = "s3://kubeedge/sedna-robo/unseen_samples/" |
| self.eval_interval = kwargs.get("eval_interval", 50) | ||
| self.no_val = kwargs.get("no_val", True) | ||
| self.cuda = True | ||
| self.savedir = '/home/QXY/dataset/save' |
There was a problem hiding this comment.
Hardcoded path should not be committed here, suggest to use the relative path.
| idp_obstacle = (x_onehot[:, 19] - tpmult[:, 19]).sum() | ||
| tp_nonobstacle = (-1*y_onehot+1).sum() | ||
|
|
||
| # for i in range(0, x.size(0)): |
There was a problem hiding this comment.
unused lines should be removed.
...ent_semantic_segmentation/lifelong_learning_bench/testalgorithms/erfnet/ERFNet/utils/loss.py
Outdated
Show resolved
Hide resolved
Signed-off-by: qxygxt <[email protected]>
Signed-off-by: qxygxt <[email protected]>
|
/lgtm |
|
/approve |
OSPP: Implementation of a Class Incremental Learning Algorithm Evaluation System based on Ianvs
No description provided.