keras-YOLOv3-mobilenet
keras-YOLOv3-mobilenet copied to clipboard
Adamdad
I transfer the backend of yolov3 into Mobilenetv1，VGG16，ResNet101 and ResNeXt101
If I want to train model with 1080p image, or other size image, what should I do? The val_loss > 4000 。。。 ` Train on 5022 samples, val on 557...
Hi, thanks for you to share this project for us. i followed this way to convert the model .h5 to .pb. [https://github.com/amir-abdi/keras_to_tensorflow ](url) but i got a problem is that...
hello,Why do vgg16 have many similar structures? For example, the following sections vgg16 = VGG16(input_tensor=inputs,weights='imagenet',include_top=False) x = vgg16.get_layer('block5_pool').output x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block6_conv1')(x) x = Conv2D(512, (3,...
I train with my own dataset with the default script my dataset has three types of objection with 40000 train sample and 6000 validation sample the batch_size is 4, my...
if input_shape is None: default_size = 224 else: if backend.image_data_format() == 'channels_first': rows = input_shape[1] cols = input_shape[2] else: rows = input_shape[0] cols = input_shape[1] if rows == cols and...
@Adamdad Hi , Conv-lstm has been added to Darknet and I am trying to convert trained yolov3+conv-lstm to .h5 using your repo (convert.py ).However, it has unsupported header error .I...
Hi, i just train a new model by feeding mobilenet pre_trained weights into mobilenet (train.py). But i got train loss non-convergence, always at 30. Should i train mobilnet first by...
Have you used the pretrained MobileNet model and construct yolo layers on top? And used pretrained MobileNet weights for Mobilenet initialization? Or you just took the architecture of MobileNet and...
Create YOLOv3 model with 9 anchors and 5 classes. [, , ] Train on 749 samples, val on 0 samples, with batch size 1. Epoch 1/300 749/749 [==============================] - 276s...