Open ardywibowo opened 3 years ago
Update: Changed the augmentations to include resized cropping to a 32x32 image size and it improved the accuracy. Will update with new results once it finishes training.
Update 2: Training has just finished and I was able to get a decent 84% accuracy with kNN. I'm still wondering if I'm missing any additional details.
Hi, could you share the resized cropping you used ? Thanks !
Sure,
It's the same as the one done in the original code: transforms.RandomResizedCrop(32, scale=(0.2, 1.))
.
For details, it may be helpful to check the demo released in the moco codebase: https://colab.research.google.com/github/facebookresearch/moco/blob/colab-notebook/colab/moco_cifar10_demo.ipynb
Hi again,
With the CIFAR ResNet18 I am getting 88.4% on knn accuracy. What were your results (on the supplementary there is not the exact number for the knn evaluation) ?
I am not sure on several things (might be details): What is the size you used for the projection head and the predictor ? In the MoCo demo they use the default arguments for RandomResizeCrop what were your values ? Also could there be a difference coming from the fact that I am working on a single GPU (and not using the trick used in the MoCo demo to split the batches) ?
Thanks :)
@elias-ramzi What did you do with your implementation? I'm only getting 84%. Did you copy the MoCo demo augmentation?
No I was use using the same augmentations as SimSiam without gaussian blur.
Did you use the ResNet version of the MoCo demo (changed conv1
, and remove maxpool
) ?
I see. No I didn't. Thanks! Let me try that.
Did you use 128 as the feature_dim
or the default 2048? If you changed it to 128, did you change the pred_dim
as well to 128/4 = 32
?
In the supplementary material it is said that they used 2048
, so I used the default settings for the projection head (with only two layers).
Hi again,
With the CIFAR ResNet18 I am getting 88.4% on knn accuracy. What were your results (on the supplementary there is not the exact number for the knn evaluation) ?
I am not sure on several things (might be details): What is the size you used for the projection head and the predictor ? In the MoCo demo they use the default arguments for RandomResizeCrop what were your values ? Also could there be a difference coming from the fact that I am working on a single GPU (and not using the trick used in the MoCo demo to split the batches) ?
Thanks :)
Hi bro, how many parameters in your CIFAR ResNet18, I can get only 78% with CIFAR resnet 18, but 89% with resnet 18
Hi, sorry for the late response:
I have 14580800 parameters for the CIFAR resnet18
hi @ardywibowo , have you tried it with tiny-imagenet dataset, I have trained and used the pre-trained model to do an linear classification but result look really bad
我在高光谱数据上跑这个代码的时候,线性分类精度只有55%,请问各位大佬,我需要在哪些地方改进,我的数据增强就是单一的高斯噪声,我的代码有问题吗,还是说我要修改优化器还是啥
您好,我正在尝试修改代码以在CIFAR-10数据集上工作,我遇到了一些麻烦。我更改了以下内容:
- LR = 0.03
- 重量衰减 = 0.0005
- 动量 = 0.9
- 批大小 = 512
- 删除了高斯模糊和裁剪
- 已删除投影仪中的第二组线性、批处理规范和 ReLU 图层。
- 已将规范化参数修改为 CIFAR-10 统计信息。
- 已将体系结构更改为 resnet18。这篇论文提到了CIFAR的ResNet18,但我不确定它是否与resnet18不同。
但我似乎无法让它工作。线性微调和 kNN 的准确率只有 37% 左右。你们是否有任何我可能错过的其他实现细节?
你好,你后面是如何提高的线性分类精度呢
我在高光谱数据上跑这个代码的时候,线性分类精度只有55%,请问各位大佬,我需要在哪些地方改进,我的数据增强就是单一的高斯噪声,我的代码有问题吗,还是说我要修改优化器还是啥
Hello, I was trying to modify the code to work on the CIFAR-10 dataset and I'm having some trouble. I have changed the following:
But I can't seem to get it to work. I'm getting only about 37% accuracy with both the Linear finetuning and kNN. Do you have any additional implementation details that I may have missed?