quark0/darts issues and pull requests

#75 - I wonder if the code is different from the algorithm in origin paper？？ thanks

Issue - State: open - Opened by QiuPaul over 5 years ago - 4 comments

#74 - concat operation for the Cell in model_search.py

Issue - State: closed - Opened by tengteng95 over 5 years ago - 5 comments

#73 - Architecture Search on Cifar100 or imagenet?

Issue - State: open - Opened by ildoonet over 5 years ago - 2 comments

#72 - How many times have you experimented for searching architectures?

Issue - State: closed - Opened by ildoonet over 5 years ago - 1 comment

#71 - what's the GPU memory size when training on cifar10 and imagenet?

Issue - State: closed - Opened by dragen1860 almost 6 years ago

#70 - Inconsistency with NASNet Search Space Cell Construction?

Issue - State: closed - Opened by mikelzc1990 almost 6 years ago - 1 comment

#69 - Derivation from Eq 5 to Eq 6 in the paper?

Issue - State: closed - Opened by dragen1860 almost 6 years ago - 1 comment

#68 - what the meaning of alphas_normal and alphas_reduce?

Issue - State: closed - Opened by raninbowlalala almost 6 years ago - 3 comments

#67 - CPU Support for those of us without GPUs

Issue - State: closed - Opened by adammenges almost 6 years ago

#66 - Using test data in train.py for validation?

Issue - State: open - Opened by dkumazaw almost 6 years ago - 1 comment

#65 - Why do NetworkCIFAR and NetworkImageNet have different stems?

Issue - State: open - Opened by e-271 almost 6 years ago - 1 comment

#64 - what's the function of `target.cuda(async=True)`?

Issue - State: closed - Opened by dragen1860 almost 6 years ago - 3 comments

#63 - Figure 3 CIFAR10 valid accuracy after 100 epochs

Issue - State: open - Opened by liamcli almost 6 years ago

#62 - Imagenet training time?

Issue - State: open - Opened by tonystark940501 almost 6 years ago - 2 comments

#61 - Problem on reproducing results of RNN on PTB

Issue - State: open - Opened by tonystark940501 almost 6 years ago - 2 comments

#60 - typical final accuracy for 1st & 2nd derivative search step?

Issue - State: open - Opened by ahundt almost 6 years ago - 2 comments

#59 - what is the purpose for the "small batch size" in RNN search?

Issue - State: open - Opened by Ian09 almost 6 years ago - 1 comment

#58 - The number of nodes in RNN cell search in the code is not consist with your paper and ENAS (baseline)

Issue - State: open - Opened by Ian09 almost 6 years ago - 1 comment

#57 - How to transfer the code to other dataset？

Issue - State: open - Opened by marsggbo almost 6 years ago - 2 comments

#56 - why in my model_search only loss.backward(retain_graph=True) can run

Issue - State: open - Opened by xiaoiker almost 6 years ago - 1 comment

#55 - How to train it on a dataset other than ImageNet?

Issue - State: closed - Opened by aminecherif94 almost 6 years ago - 2 comments

#54 - Is it ok to copy the 49th epoch's genotype?

Issue - State: open - Opened by WayneZww almost 6 years ago - 1 comment

#53 - about _compute_unrolled_model

Issue - State: open - Opened by xiaoiker almost 6 years ago - 4 comments

#52 - architect.py

Issue - State: closed - Opened by SivanDoveh about 6 years ago - 1 comment

#51 - I get an error when I want to change the stirde

Issue - State: open - Opened by zhangdan94 about 6 years ago - 2 comments

#50 - Poor PTB test performance?

Issue - State: closed - Opened by D-X-Y about 6 years ago - 12 comments

#49 - Train_imagenet setting

Issue - State: closed - Opened by ashaw596 about 6 years ago - 2 comments

#48 - FC/Dense and MLP capabilities

Issue - State: closed - Opened by jaggzh about 6 years ago - 1 comment

#47 - Validation phase code

Issue - State: closed - Opened by vardaan123 about 6 years ago - 2 comments

#47 - Validation phase code

Issue - State: closed - Opened by vardaan123 about 6 years ago - 2 comments

#46 - The code to calculate the multiply-add operations?

Issue - State: closed - Opened by D-X-Y about 6 years ago - 11 comments