keras-yolo3

keras-yolo3 copied to clipboard

Training on Coco dataset from scratch

3

Hi, Has anyone been able to train on the coco data from scratch successfully? When I am training the train loss stops improving at 60 and I get an mAP...

guskarls

Loss lambda parameters

1

Hi! I've realized that this loss implementation doesn't include the λcoord and λnoobj parameters. Is there any reason for that? Does the original Yolo v3 use them?  Thanks!

AlbertoSabater

is there any wrong about the confidence loss?

1

origin: **confidence_loss = object_mask * K.binary_crossentropy(object_mask, raw_pred[...,4:5], from_logits=True)+ (1-object_mask) * K.binary_crossentropy(object_mask, raw_pred[...,4:5], from_logits=True) * ignore_mask** i think , it means the different cross entroypy value of labels = 1 and...

ztfmars

大哥们解决了吗？我是用model.save()保存，也是打算转化为pb模型。。。但是load_model()就遇到了NameError: name 'yolo_head' is not defined的问题。。。有没有样例给我看看啊，或者指个方向，谢谢

3

大哥们解决了吗？我是用model.save()保存，也是打算转化为pb模型。。。但是load_model()就遇到了NameError: name 'yolo_head' is not defined的问题。。。有没有样例给我看看啊，或者指个方向，谢谢 _Originally posted by @Rainweic in https://github.com/qqwweee/keras-yolo3/issues/349#issuecomment-515644677sa _

AadeIT

the pretrained model was trained on what dataset

1

hello sir i wonder the pretrained model was trained on what dataset

yuningkang

the value of accuracy is too big

1

Hello! After I running kmeans.py,the result of the accuracy is too big. K anchors: [[ 82 81] [163 248] [186 429] [241 347] [271 416] [313 460] [340 324] [390...

tangbinbinyes

Feature Request: do the same to tiny-yolo-v4

3

ybloch

converting trained model doesn't work, while the pretrained model works..?(coreML)

7

hey! thanks! I successfully trained my first model, but after that I am not able to convert it using: " import coremltools coreml_model = coremltools.converters.keras.convert('model_data/trained_weights_final.h5', input_names='input1', image_input_names='input1', output_names=['output1', 'output2', 'output3'],...

JcbPrn

Updated the format using black which is a standard format. Do merge it if satisfactory

parikshit14

no code to get maP for results

1

how can i get maP (mean average precision) for my trained dataset?

faheem-khaskheli