dragen1860/MAML-Pytorch
Elegant PyTorch implementation of paper Model-Agnostic Meta-Learning (MAML)
PythonMIT
Issues
- 0
- 0
训练自己的数据集
#76 opened by qiongjiekucha - 0
准确率不变
#75 opened by shiyao1999 - 0
- 6
About accuracy
#73 opened by feisichen - 1
Why the code for Learner is so complicated?
#69 opened by fikry102 - 3
您好,对于代码有两个问题,请教您一下,谢谢
#65 opened by CUITCHENSIYU - 2
- 1
请问模型权值文件在哪里进行保存?代码在哪里?
#70 opened by XingshiXu - 2
Should I fine-tune the model on each new task
#30 opened by AaboutL - 0
- 6
About training and testing
#66 opened by zkk-web - 0
debug
#71 opened by swaggyP9527 - 6
- 1
omniglot dataset download error
#58 opened by vegetarianfish - 1
- 3
- 0
Asking about inner and outer loop
#64 opened by tranquangchung - 3
Incorrect losses_q
#61 opened by zilin129 - 0
about dataset spiltting
#62 opened by raymond00000 - 1
- 0
- 1
Does the hessian really gets computed?
#59 opened by tanghl1994 - 1
questions with respect to the no_grad()
#40 opened by d12306 - 2
Some ideas about parameters updating.
#47 opened by zhaoyu-li - 0
why use custom grad clip function?
#57 opened by QasimWani - 3
- 1
Overlap between meta-training and meta-testing?
#56 opened by atseng17 - 17
Measure the model performance
#42 opened by kk2491 - 3
- 5
run code have a mistakes on db = DataLoader(mini, args.task_num, shuffle=True, num_workers=1, pin_memory=True)
#33 opened by hongshengxin - 2
2nd Order or 1st Order Approximation?
#32 opened by Vampire-Vx - 0
something about Computational Graph
#54 opened by kongbia - 0
how to save and load the model
#53 opened by asker-github - 1
Question about variable, losses_q
#52 opened by sailist - 2
Why not "loss_q = sum(losses_q) / task_num" ?
#51 opened by geekyutao - 0
What's the difference between task_num and N-way?
#49 opened by luciaL - 1
accuracy format
#41 opened by eghouti - 1
Only use Conv and Linear
#46 opened by CongWeilin - 1
- 1
RuntimeError: Expected object of scalar type Long but got scalar type Int for argument #2 'target'
#45 opened by Peterdingpeng - 0
- 1
Conv-ReLU-BN issue
#39 opened by mileyan - 2
why not set bn_training=False when test net on x_qry in finetunning phase
#27 opened by boy-be-ambitious - 0
Some questions about the codes
#43 opened by HongduanTian - 1
OmniglotNShot
#37 opened by li829 - 0
Mutiple GPU
#38 opened by dipanjan06 - 0
learning_rate
#31 opened by dawn2034 - 1
Maybe there is some error in learner.py?
#29 opened by crashmoon - 0