Training Custom Dataset using tiny-yolo-v4

[rak7045]

Hello @Tianxiaomo, I am trying to train my custom dataset using tiny-yolo-v4. I have made changes in the .cfg file and trying to train the network. The following error has been raised. Do you have any idea on this? Could you please help me to sort this?

For clarification, I have used weights file downloaded from Yolo website, since you do not have the weights file for tiny yolov4. I have given train.txt path and path for images as you mentioned. While training I got error such as "ValueError: string could not be converted into float". I have written this piece of code where the error is resolved.(Line 258 in dataset.py)

data = line.strip().replace("\n","").replace("\r", "").replace(" ", "$")

print(data)

        #data = data.replace(",$", ",")
        data = data.replace("$", " ")
        data = data.split(" ")
        truth[data[0]] = []
        for i in data[1:]:
            truth[data[0]].append([float(j) for j in i.split(',')])  (Might be helpful to someone)

[pavlik-tikhomirov]

Hello @Tianxiaomo, I am trying to train my custom dataset using tiny-yolo-v4. I have made changes in the .cfg file and trying to train the network. The following error has been raised. Do you have any idea on this? Could you please help me to sort this?

For clarification, I have used weights file downloaded from Yolo website, since you do not have the weights file for tiny yolov4. I have given train.txt path and path for images as you mentioned. While training I got error such as "ValueError: string could not be converted into float". I have written this piece of code where the error is resolved.(Line 258 in dataset.py)

data = line.strip().replace("\n","").replace("\r", "").replace(" ", "$") #print(data) #data = data.replace(",$", ",") data = data.replace("$", " ") data = data.split(" ") truth[data[0]] = [] for i in data[1:]: truth[data[0]].append([float(j) for j in i.split(',')]) (Might be helpful to someone)

turn strides in train.py around in the opposite direction