• Tensorflow transfer learning EfficientNet on Stanford dogs

  • By: tensorflow
  • Podcast

Tensorflow transfer learning EfficientNet on Stanford dogs

By: tensorflow
Listen for free

  • Summary

  • in Mandarin/zh, with EfficientNet, with pretrained weights, freeze all the weights, then you can add your top, then fit/train, you can reach satisfied accuracy with very limited training time/epochs; then unfreeze the layers -20: of non BN layers, change its trainable to True, then fit, you can get even better accuracy with very few like 10 epochs. So this is called transfer learning.
    tensorflow
    Show more Show less
Mathematics Science
Show more Show less
activate_Holiday_promo_in_buybox_DT_T2
Episodes
  • OCR with ctc_layer to read captcha

    OCR with ctc_layer to read captcha
    Sep 11 2023

    OCR with ctc_loss_layer to read captchas

    带ctc损失层的keras OCR模型来读取网页上的captcha;

    训练模型最后是ctc_layer, 训练后模型不能直接用于推断，另外建立推断模型，使用原基础模型的输入 和 最后Dense层的输出 而建立新模型。推断结果也需要进行额外的解析
    Show more Show less
    1 hr and 9 mins
  • keras Image Classification, cats and dogs 图片分类识别 猫和狗

    keras Image Classification, cats and dogs 图片分类识别 猫和狗
    Aug 10 2023

    in Mandarin/zh, the simplest way, keras Image Classification, build a model to classify cats and dogs, total two classes, that is, cat, dog. You download pics dataset of cats and dogs from microsoft, then apply augmentation ( translate, rotate, flip, contrast), build a residual convo model, compile and fit, save the model or keep the weights, now you can predict / inference if it is a cat or a dog. I took some pics of cats and dogs in the yard, nnnn, the result is not so good, I mean the accuracy is bad. Perhaps I shall let the cat/dog take major surface ratio like 80 -90% when shooting pics.
    Show more Show less
    1 hr and 11 mins
  • transfer learning, use EfficientNet fine-tune to fit Stanford Dogs, in Mandarin

    transfer learning, use EfficientNet fine-tune to fit Stanford Dogs, in Mandarin
    Aug 8 2023

    transfer learning in Mandarin: 迁移学习，用预训练好的权重， EfficientNet, 不要top, 只有权重，冻结权重， 再建立自己的top, 训练后， 再把 后20层 非BN层的 解冻，改为可以训练权重， 再训练10个epo, 达到令人满意的准确度，而耗用的资源/计算力 非常有限
    Show more Show less
    1 hr and 15 mins

What listeners say about Tensorflow transfer learning EfficientNet on Stanford dogs

Average customer ratings

Reviews - Please select the tabs below to change the source of reviews.

Audible.com reviews

Amazon reviews
No Reviews are Available