Stack Overflow
Stack Overflow

Questions tagged [mobilenet]

205 questions
2
votes
0 answers

'Channels first' training accuracy very low compared to 'channels last'

My issue: I am trying to train a semantic segmentation model in tf.keras, in fact it works very well when I am using channels_last (WHC) mode (it reaches 96%+ val acc). I wanted to train it in channels_first (CHW) mode so the weights are compatible…
Neil Marko
  • 33
  • 1
  • 6
2
votes
1 answer

ssd_mobilenet_v2_coco can't detect custom trained objects after exporting inference graph

I'm trying to detect marigolds on a field using the tensorflow api. I have already tried it with the faster_rcnn_inception model which worked pretty well but i'm planning to run the detection on raspi4 and for that it's too heavy. So I tried…
simon
  • 23
  • 5
2
votes
1 answer

Image Classification Graph model making wrong predictions

I'm using the make_image_classifier python script to retrain a mobilenetv2 on a new set of images. My end goal is to make predictions in tfjs in the browser. This is exactly what i'm doing: Step 1: Retrain the model make_image_classifier \ …
Antoine Vincent Jebara
  • 23
  • 2
2
votes
2 answers

Tensorflow object detection API not working even loss is low

I want to create a model with tensorflow object detection API to detect card numbers in credit cards. So I prepare my dataset of cards about 50000 cards for training and 15000 cards for validation.My model is SSD_Mobilenet_V1_0.25_224 and I run…
Shayan Tabatabaee
  • 448
  • 4
  • 14
2
votes
0 answers

ValueError: Tensor Conversion Error Tensorflow

I create tfrecord files and put them in the right directory C:\Users\Admin\Desktop\models\research\object_detection\data and then I run train.py then this error pops up Traceback (most recent call last): File "legacy/train.py", line 184, in…
jsn-bln
  • 63
  • 1
  • 7
2
votes
0 answers

Is there any keras code to reproduce the weight pruning of MobileNet?

I want to prune the MobileNetV1 model using the weight pruning method from https://www.tensorflow.org/model_optimization/guide/pruning However, I don't know how to set the pruning_schedule. pruning_params = { 'pruning_schedule':…
bin.li
  • 61
  • 5
2
votes
1 answer

How to add a few layers before the model in transfer learning with tensorflow

I am trying to use transfer learning in tensorflow. I know the high level paradigm base_model=MobileNet(weights='imagenet',include_top=False) #imports the mobilenet model and discards the last 1000 neuron…
figs_and_nuts
  • 4,870
  • 2
  • 31
  • 56
2
votes
1 answer

Cannot serialize protocol buffer when using MobileNet with Tensorflow Federated

I'm using the pre-trained MobileNet from Keras and want to train it using TensorFlows federated learning, but I'm always getting an error that the protocol buffer cannot be serialized since the 2GB limit is exceeded. My inputs are 224x224 RGB…
Recep26
  • 175
  • 1
  • 9
1
vote
0 answers

Figure of `inverted residual with linear bottleneck` shown in mobilenetV2 paper is compliant with MobileNetV2 implementation of `Pytorch`

The MobileNet V2 paper introduced inverted residual with linear bottleneck block. It is shown diagrammatically in Fig. 3(b) (Same is shown in Fig. 3 of MobileNetV3 paper) and mathematically in Table 1 of this paper. Fig. 3(b) shows that the…
Palguna Gopireddy
  • 19
  • 6
1
vote
0 answers

Failed to evaluate CNN Model (MobileNet V3) with eval.py

I built a Machine Learning Model with Tensorflow CNN (MobileNetV3), with around 1000 images (one image contains multiple objects), 80 classes, and with 1024x1024 pixels per image. I was using 50000 epoch to train the model. When I finished the…
Ahmad Azzam Alhanafi
  • 11
  • 3
1
vote
2 answers

Why my model training automatically stopped during training?

I was trying to make a custom detector using SSD MobileNet. I never tried this model before for making a custom detector using TensorFlow API. I was following this tutorial. I am facing a problem after the start of my model training, where the…
Md Mahadi Hasan Sany
  • 129
  • 1
  • 1
  • 12
1
vote
0 answers

ValueError: Shapes (None, None) and (None, 8, 2, 19) are incompatible

I have changed input size many times but i cant find the issue in my mobilenet v1 224. # MobileNet block def mobilnet_block (x, filters, strides): x = DepthwiseConv2D(kernel_size = 3, strides = strides, padding = 'same')(x) x =…
Rubi Choudhary
  • 11
  • 1
1
vote
0 answers

Getting Nan value for the loss while training a SSD-Mobilenet that was using a custom loss function

Implementing the SSD-Mobilenetv1 from scratch with Pascal VOC -2007 and 2012 dataset and using GPU 3060 for training in tensorflow mainly using this code - https://github.com/ManishSoni1908/Mobilenet-ssd-keras. the error would look like Epoch 1/10 …
manish
  • 21
  • 1
1
vote
3 answers

Keras model returns high validation accuracy while training, but accuracy is very low while evaluating

I am trying to train a simple MobileNetV3Small under keras.applications as shown below base_model = keras.applications.MobileNetV3Small( input_shape= INPUT_SHAPE, alpha=.125, include_top=False, classes=1, …
Bhoke
  • 468
  • 6
  • 22
1
vote
1 answer

Tensorflow lite object detection, Android Studio, ssd mobilevet v2, same structure different tflite file but almost 0 detection

I want to make object detection application base on this github https://github.com/bendahouwael/Vehicle-Detection-App-Android. That github code uses tflite based on ssd mobilenet v1. So I made my custom model based on ssd mobilenet v2. I followed…
ywseo97
  • 51
  • 3
1 2
3
…
13 14