-
Each class is divided into 1300 images for training and 50 images for validation.The data sets I found online only have 1300 training sets per class, how do I get the test sets
-
https://arxiv.org/abs/2303.13664 shows that varying the temperature in the contrastive loss during pretraining can help models learn better representations for datasets with long-tail distributions an…
-
Hi,
I am trying to evaluate the mcunet-320kB implemented in TFLite using your eval_tflite.py script. I followed the steps provided in this Git issue ( https://github.com/mit-han-lab/mcunet/issues/1…
-
Thanks for sharing the code.
I have the imagenet2012 dataset, which is more than 100GB. How can I process it to 6.1GB preprocessed dataset.
Looking forward to your reply~
-
thanks for this great job. your paper just report the minist\cifar scores. do you have train model for imagenet or other largescale dataset? can you share more experience for this largescale image …
zjupx updated
5 years ago
-
Hi @GengDavid ,
Currently, I am trying to replicate the results of the SLCA approach. I have a question regarding the ImageNet-R dataset and would appreciate your assistance.
In the paper, it i…
-
hi @kaiyuyue
i see the [lines](https://github.com/kaiyuyue/nxtp/blob/main/src/encoding.py#L115-L119) in `encoding.py`, for what reason to miss these samples? for imagenet-1k dataset, many categories…
-
### Describe the bug
I'd expect the following code to download just the validation split but instead I get all data on my disk (train, test and validation splits)
`
from datasets import load_data…
-
Your work is exceptionally fascinating and highly commendable. Could you kindly clarify which version of the ImageNet dataset you utilized in your experiments? For instance, was it the classic 2012 ve…
6akso updated
1 month ago
-
### Version
1.40.0.dev20240628
### Describe the bug.
Hi,
I am testing the throughput of DALI (v1.40.0.dev20240628) when reading TFRecords via S3 and I am obtaining unexpectedly poor results.
…