neerajj9 / text-detection-using-yolo-algorithm-in-keras-tensorflow Goto Github PK

Hello. Thank you for your model.
I have faced with a problem. I don't have any accuracy improvements during training, the loss is also hovering at the some value, with the accuracy and loss being in the vicinity of 0.058 and 8.75 respectively

I have tried to use several optimizers:

1. Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.0)
2. keras.optimizers.RMSprop(lr = 0.0001)
3. keras.optimizers.RMSprop(lr = 0.001)

and I try to change a batch size :

1. 4
2. 8
3. 16

Actually, no effect
Could you help me to tackle this problem?

Hi Neeraj,

I am getting huge X.npy file after pre-processing. For the images which are of total 248 MB, i am getting X.npy file of size 2.8GB.
I am using your same data set only.
I am just a beginner into this domain..
Can you please let me know where I am doing it wrong or is it okay to have this huge file?
Are u also getting these huge npy files?
Please help me here..
Thanks in advance
Suvajit

@Neerajj9

Hello,
I'm trying to fit the model with dataset that u linked on readme. During my test loss value starts at 200 - 800 (depend on batch size and lr) and go to infinite in 2-4 steps.

Hope to receive soon a feedback,
best regards

i have copied this code and run it. It show me this error. What should i do?

e.g a 800 * 1200 picture

where is text_detect.h5 file in model directory?

I have downloaded your pre-trained model and tested it by training 1 epoch with the same dataset, but the accuracy was low 0.25 and val_loss was at about 3.5. Was it normal?