yolov3-tf2

yolov3-tf2 copied to clipboard

I couldn't understand this piece of code.

1

```python # 3. inverting the pred box equations grid_size = tf.shape(y_true)[1] grid = tf.meshgrid(tf.range(grid_size), tf.range(grid_size)) grid = tf.expand_dims(tf.stack(grid, axis=-1), axis=2) true_xy = true_xy * tf.cast(grid_size, tf.float32) - \ tf.cast(grid, tf.float32)...

ramsane

Difference in output from saved weights or saved model

2

I tested Tiny YOLOv3 on a video using weights from training as well as a saved model created using `tools/export_tfserving.py`. The outputs can be seen [here](https://imgur.com/a/pnzaRgf). In the saved model...

favre49

Hi, I'm trying to train the model on a modified version of the xView dataset which leads me to some questions. 1. I would like to add some negative examples...

JonaGanz

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

15

I am currently training this on the xview dataset for just 1 class. To note, each image is around 3000^2 pixels and each object in the image is super small...

Hsengiv2000

How to resume training

4

How do i resume training, what changes shiuld i make to continue training from where it left

abhishekbu

训练过程中出现问题:

14

指令：python train.py --dataset ./data/Face/face_train.tfrecord --val_dataset ./data/Face/face_val.tfrecord --classes ./data/Face/face_label.names --num_classes 2 --mode fit --transfer darknet --batch_size 2 --epochs 10 --weights ./checkpoints/yolov3.tf --weights_num_classes 80 我的数据集通过voc_2012.py变为tfrecord格式，用visual_dataset.py也能从中随机抽取图片，并精确的标注出了目标，但是每当我训练的时候，到了第420次时，总出现以下问题，请问有谁知道问题出在哪里吗？ 2020-03-17 22:47:23.177447: W tensorflow/core/common_runtime/bfc_allocator.cc:305] Garbage collection: deallocate...

Eve66666

hi i tested this on Raspberry Pi

3

got 4 fps with yolov3-tiny on detect-video from webcam

funkydude755

我尝试对一个物体进行训练（火焰），loss降低到很低，但是使用模型没有任何输出

1

tsk15535904190

ValueError: cannot reshape array of size 1518 into shape (33,256,1,1)

3

While executing `python convert.py --weights ./data/yolov3-tiny.weights --num_classes 6 --output ./checkpoints/yolov3-tiny.tf --tiny` to convert my tiny yolo weights i am having this issue:  please help

VeliAliyev

Real y_true structure?

2

I'm trying to figure out the right structure of data being fed to the model during training: In train.py, you map dataset.py's transform_targets() fn, which in turn applies transform_targets_for_output(), which...

rachelthorp