Stack Overflow
Stack Overflow

Questions tagged [efficientnet]

questions about retraining efficientnet with Keras to create new object detection models

efficientnet is a lightweight convolutional neural network architecture achieving the state-of-the-art accuracy with an order of magnitude fewer parameters and FLOPS, on both ImageNet and five other commonly used transfer learning datasets.

109 questions
0
votes
0 answers

Internal Error: EagerConst: Dst tensor is not initialized

I have tried to use efficient net as shown below but before training model throws an error as mentioned after the code from tensorflow.keras import layers from tensorflow.keras.applications import EfficientNetB0 NUM_CLASSES = 5 IMG_SIZE = 224 size…
Shradha Dubey
  • 1
0
votes
0 answers

Keras - EffNetB3 and B6 and ResNet50 - Out of Memory issue when trying to unfreeze model and fine-tune

I'm trying to learn Image classification and have got some models working on datasets that I have generated. I've used ResNet50, EfficientNetB3 and B6 for transfer learning and I have a working model at the end but would like to improve their…
Chris
  • 1
0
votes
0 answers

loading pretrained efficientnet from NVIDIA stopped working

I used to load a base Efficientnet classifier like this : model = torch.hub.load('NVIDIA/DeepLearningExamples:torchhub', 'nvidia_efficientnet', type='efficientnet-widese-b0') Recently this stopped working. I now get : Downloading:…
Alix Martin
  • 332
  • 1
  • 5
0
votes
0 answers

Efficientnet normalization layer

Question 1 First 2 layers of efficientnet are Rescaling and Normalization Layers. I was reading how the normalization layer works and before normalization, adapt() function should be called. I looked efficientnet source code and there was no…
bora eruysaliz
  • 1
  • 1
0
votes
1 answer

RuntimeError: Expected number of channels in input to be divisible by num_groups, but got input of shape [64, 16, 32, 32] and num_groups=32

I have EfficientNet working fine on my dataset. Now, I changed all the batch norm layers into group norm layers. I have already done this process with other networks like vgg16 and resnet18 and all was ok. On EfficientNet I have this error…
CasellaJr
  • 378
  • 2
  • 11
  • 26
0
votes
2 answers

ValueError: Shapes (None, None) and (None, None, None, 3) are incompatible

I am using EfficientNetV2B0, everything works perfectly until when I try to fit the model. On the train_generator I tried the class_mode = 'sparse', and also class_mode = 'categorical', it throws different kinds of error messages. It took me few…
Bet
  • 19
  • 2
0
votes
1 answer

Data type preference for training CNN?

I originally was using input data of int8 type ranging from 0-255 before learning that standardizing and normalizing should increase learning speeds and accuracy. I attempted both, with and without a mean of zero, and none of these methods improved…
numunu
  • 25
  • 5
0
votes
1 answer

How to use a CNN code in python inside a website?

I have website with backend in Python (Django) and JavaScript hosted on heroku. Also, I have code in python that does image classification with EfficientNet, so I want to integrate this code into my website. The logical sequence of ideas is as…
Diego Freire
  • 40
  • 6
0
votes
0 answers

Testing Concept Activation Vector (TCAV) for Pytorch

I am trying to compute the Testing Concept Activation Vectors (TCAV, as described here) vectors for different concepts for my classification model. So far, I haven't successfully found code online for Pytorch models so I have decided to rewrite it…
rkraaijveld
  • 23
  • 4
0
votes
1 answer

I want to ask about this error sbouy input shape, i'm making a thesis about image classification using efficientnet

conv_base = EfficientNetB6(weights="imagenet", include_top=False, input_shape=400,400) SyntaxError: positional argument follows keyword argument #Options: EfficientNetB0, EfficientNetB1, EfficientNetB2, EfficientNetB3, ... up to 7 #Higher the…
Mark Stevan
  • 1
  • 1
0
votes
1 answer

load recent checkpoints with efficientnet

I try to do transfer learning to efficientnet in tensorflow. I want to use noisy-student checkpoints instead of imagenet weights: model = EfficientNetB3(weights='noisy_student_efficientnet-b3', include_top=False,…
Shirly
  • 38
  • 6
0
votes
1 answer

How EfficientNet width scaling affect the FLOPS of the model?

I'm re-building the EfficientNet by Tan et al. (2019). And there's something I don't understand in the width scaling factor. In the paper they tried to maximize the model accuracy for given resource constraints as: Based on the above equation, the…
Do Nam
  • 11
  • 4
0
votes
1 answer

Why does my validation loss / accuracy fluctuate despite manual test show good results

I am training an EfficientNet lite (from scratch) on a dataset of ~10.000.000 images (128x128x1) with ~6500 classes. My training loss is converging as well as my training accuracy. However, my test loss/accuracy are fluctuating. When I test the CNN…
CaptainDario
  • 83
  • 1
  • 7
0
votes
1 answer

Unable to Import Efficientnet in Colab

I am working in colab to test a code. While importing models, its giving error No module named 'efficientnet' I am sharing the code and error here. # for accessing tabular data import pandas as pd import numpy as np import…
Data_Science_Learner
  • 19
  • 6
0
votes
1 answer

openvino human pose estimation 0007

I am working with openvinotoolkit and I am trying to inference the model human-pose-estimation-0007 and that works fine on neural stick 2. Now I am looking for the original human-pose-estimation-0007 weights o repository (Multi-person 2D pose…
ssgakhal
  • 408
  • 3
  • 15
1 2 3 4
5
6 7 8