dragen1860 / MAML-Pytorch

Elegant PyTorch implementation of paper Model-Agnostic Meta-Learning (MAML)

dragen1860/MAML-Pytorch Issues

准确率不变
Updated 5 months ago4
can you offer us your requirements of environment?
Closed 5 months ago
训练自己的数据集
Closed 8 months ago
What is the backup file for and what is the reference navie5 in navie5?
Updated a year ago
About accuracy
Updated a year ago6
Why the code for Learner is so complicated?
Updated 2 years ago1
您好，对于代码有两个问题，请教您一下，谢谢
Updated 2 years ago3
Why is `for epoch in range(args.epoch // 10000):`
Updated 2 years ago2
请问模型权值文件在哪里进行保存？代码在哪里？
Updated 2 years ago1
Should I fine-tune the model on each new task
Closed 5 years ago2
Using the Learner object for my project, Loss not behaving at its best
Updated 2 years ago
About training and testing
Updated 2 years ago6
debug
Updated 2 years ago
How to use the trained MAML model for predicting one image at a time?
Updated 2 years ago6
omniglot dataset download error
Closed 4 years ago1
create_graph parameter is False hence first-order MAML?
Updated 3 years ago1
Performance on Omniglot is slightly lower than paper report
Updated 3 years ago3
Asking about inner and outer loop
Updated 3 years ago
Incorrect losses_q
Updated 3 years ago3
about dataset spiltting
Updated 3 years ago
Something wrong in the config of the last max pooling layer.
Updated 3 years ago1
Can you please add a 1-d CNN model to the learner.
Updated 4 years ago
Does the hessian really gets computed?
Updated 4 years ago1
questions with respect to the no_grad()
Updated 4 years ago1
Some ideas about parameters updating.
Updated 4 years ago2
why use custom grad clip function?
Updated 4 years ago
Result is not deterministic because of iterating over Python dictionary
Updated 4 years ago3
Overlap between meta-training and meta-testing?
Closed 4 years ago1
Measure the model performance
Updated 4 years ago17
Re-sampling tasks after each epoch increases the performance
Updated 4 years ago3
run code have a mistakes on db = DataLoader(mini, args.task_num, shuffle=True, num_workers=1, pin_memory=True)
Updated 4 years ago5
2nd Order or 1st Order Approximation?
Updated 4 years ago2
something about Computational Graph
Updated 4 years ago
how to save and load the model
Closed 4 years ago
Question about variable, losses_q
Closed 4 years ago1
Why not "loss_q = sum(losses_q) / task_num" ?
Closed 4 years ago2
What's the difference between task_num and N-way?
Updated 5 years ago
accuracy format
Updated 5 years ago1
Only use Conv and Linear
Updated 5 years ago1
for step, (x_spt, y_spt, x_qry, y_qry) in enumerate(db):
Updated 5 years ago1
RuntimeError: Expected object of scalar type Long but got scalar type Int for argument #2 'target'
Updated 5 years ago1
why running_mean/running_var: requires_grad must be False?
Updated 5 years ago
Conv-ReLU-BN issue
Updated 5 years ago1
why not set bn_training=False when test net on x_qry in finetunning phase
Updated 5 years ago2
Some questions about the codes
Updated 5 years ago
OmniglotNShot
Updated 5 years ago1
Mutiple GPU
Updated 5 years ago
learning_rate
Updated 5 years ago
Maybe there is some error in learner.py?
Closed 5 years ago1
Confused
Closed 5 years ago