sunshineatnoon / darknet.keras Goto Github PK

I run the TrainYolo.py and using Tensorflow as keras backend, howerver i get exception:
Exception: Error when checking model target: expected dense_3 to have shape (None, 1470) but got array with shape (16, 1225)
the shape of label and the output of the layer is different.

How to fine tune your model?
I don't have sufficient data to retrain your model from scratch.
I want to fine tune your model on my data which has only two classes ?

Hi, I have downloaded yolo-tiny.weights and put it into folder 'weights'.

Then happens an ERROR:
Using Theano backend.
Traceback (most recent call last):
File "RunTinyYOLO.py", line 194, in
yoloNet = ReadTinyYOLONetWeights(os.path.join(os.getcwd(),'weights/yolo-tiny.weights'))
File "D:\DeepLearningProgram\Darknet.keras-master\utils\TinyYoloNet.py", line 80, in ReadTinyYOLONetWeights
l.biases = np.asarray(testArray[0][count])
IndexError: index 0 is out of bounds for axis 0 with size 0

And I check file 'TinyYoloNet.py', def 'ReadTinyYOLONetWeights', the 'testArray' variable is [] (null) with many many dtype.

What is the problem ??

Hi, there are supposed to be 1470 parameters in the last layer of Yolo-Tiny. But when I do a model.summary(), there are 1225 parameter, which is giving me an out of bounds error when I try to do detection with it.

I followed the instructions but I can not run code
I got the following
line 80, in ReadTinyYOLONetWeights
l.biases = np.asarray(testArray[0][count])
IndexError: index 0 is out of bounds for axis 0 with size 0

@sunshineatnoon Hi, could you please report your average runtime per each image when you run the yolo-tiny network? And which graphic card did you use on it? Thanks !!

How can we perform training using Keras?

Dear @sunshineatnoon your repo is really cool,
after using the crop function in utils, I tried to feed the image into the net I built in tensorflow as follows:

class SimpleNet(object):
    def __init__(self, yolo):

        self.input_x = tf.placeholder(tf.float32, 
            [None, 448, 448, 3], name = "input_x")

        now = self.input_x
        for i in range(yolo.layer_number):
            l = yolo.layers[i]
            if (l.type == "CONVOLUTIONAL"):
                now = tf.nn.conv2d(
                    now,
                    tf.Variable(l.weights.transpose([2,3,1,0])),
                    strides = [1,1,1,1],
                    padding = "SAME")
                now = tf.nn.bias_add(now, tf.Variable(l.biases))
                now = tf.maximum(0.1 * now, now)
            elif l.type == 'MAXPOOL':
                now = tf.nn.max_pool(now,
                    ksize = [1,2,2,1],
                    strides = [1,2,2,1],
                    padding = "SAME")
            elif l.type == 'FLATTEN':
                now = tf.reshape(now, 
                    [-1, int(np.prod(now.get_shape()[1:]))])
                print ('add flat')
            elif l.type == 'CONNECTED':
                now = tf.nn.xw_plus_b(now,
                    tf.Variable(l.weights),
                    tf.Variable(l.biases))
            elif l.type == 'LEAKY':
                now = tf.maximum(0.1 * now, now)
            elif l.type == 'DROPOUT':
                pass
        self.out = now

And then try to feed the image into the net as follows:

                image = crop(img,resize_width=512,resize_height=512,new_width=448,new_height=448)
        image = image.transpose([1,2,0])
        image = np.expand_dims(image, axis = 0)
        feed_dict ={
            model.input_x: image
        }
        timer = Timer()
        timer.tic()
        out = sess.run([model.out], feed_dict)[0]

Can you help me why is it that for the same input image our nets produce different results (out != predictions) and yours seem to be correct one, I have double checked the net construction in tensorflow above and cannot find any mis-correspondence between our nets.

Thank you.

return empty list from np.fromfile at TinyYoloNet.py, ReadTinyYOLONetWeights. line 76

how to fix it?

I'm not sure what changes have to be made to ReadTinyYOLONetWeights to use the updated weights. Can I still download the old weights? Or what changes do I have to make to the function?