Stack Overflow
Stack Overflow

Questions tagged [custom-training]

90 questions
0
votes
0 answers

Using 3DCNN for images classification, not video

I want to use 3DCNN for images classification, i.e. classification with a sequence of images as inputs to 3DCNN models, then output the corresponding feature maps, during the training, how to define the loss function and accuracy? I implemented a…
Fanghui Chen
  • 1
0
votes
0 answers

Can't iterate over `y_true` and `y_pred` in custom TensorFlow loss function

I wrote a custom TensorFlow loss class, called CustomLoss. The CustomLoss class has 2 methods: the first one is loss which calculates loss for a single object, second one is call which calls loss method on every model output. As far as I know call…
iamSrte
  • 1
  • 2
0
votes
1 answer

Loss and CER increases when training a fine-tuned TrOCR on custom dataset

I want to train the TrOCR on my custom dataset of receipts. Since we are to use the OCR for receipts we chose the “printed” fine-tuned model. We use a dataset of 5000 bounding boxes where each contains a word. However we experience that all metrics…
SauceyP
  • 1
  • 1
0
votes
0 answers

Training with output of intermediate layers in Tensorflow 2 for Soft Nearest Neighbour loss

I would like to train with the output of intermediate layers to implement the soft nearest neighbour as a regularizer as stated in https://arxiv.org/pdf/1902.01889.pdf. Therefore, I have try to implement this via a Gradient tape. with…
Trevor Yap
  • 1
  • 1
0
votes
0 answers

Mobile SSD models are expected to have exactly 4 outputs, found 8

**Im using sagemaker to train the data it has pretrained model ** -- “tensorflow-od1-ssd-resnet50-v1-fpn-640x640-coco17-tpu-8” **For Training: ** Create the SageMaker model instance. Note that we need to pass Predictor class when we deploy model…
ML Begginer
  • 1
0
votes
1 answer

yolov8: Can I add more images or data set to the model already done custom training?

I'm trying to make image detection model with 500 images. I just tried 128 epochs which took about 3 hours of training. I like the way it is working but I want to train more to get close to perfect because it has some difficult time to recognize…
Muto
  • 1
  • 3
0
votes
0 answers

Tensorflow/Keras model retraining update only non-zero weights

I have a pruned TF model which I need to retrain with the streaming data. Model includes an embedding layer (10 nodes), GRU layer (16 notes) and classification layer (85 nodes). I want to retrain only the non-zero weights of the 80% pruned…
Sampath Rajapaksha
  • 111
  • 1
  • 1
  • 11
0
votes
0 answers

TypeError: len() of a 0-d tensor while training Faster RCNN model

I’m trying to train a Faster RCNN object detection model but im getting this error. Why should it be? Error line in dataset.py: # apply the image transforms if self.transforms: sample = self.transforms(image =…
lawlicat17
  • 13
  • 3
0
votes
0 answers

Saving the IDs for cars with the help of Centroid, The IDs don't remove or update

I am using OpenCv and Trained model and trying assign IDs to the cars with the help of Centroid. There are the following scenarios: at the beginning all recognized cars get centroid and ID the cars that leave the frame, their IDs should be…
invisible
  • 3
  • 2
0
votes
0 answers

required libraries works only from second time on Сolab when using TF2 for mask rsss

could you please help me with following problem: I used TF2 for MaskRCNN. The first cell gives such error. I did it looking here first cell !apt-get update !pip3 install scikit-image==0.16.2 !pip3 install opencv-python !pip3 install…
Fatikha Work3
  • 1
0
votes
1 answer

Millipede @ Codewars

I am practicing with codewars and started doing my first challenge that is about the following: The set of words is given. Words are joined if the last letter of one word and the first letter of another word are the same. Return true if all words of…
Luis Lopez
  • 1
0
votes
0 answers

Tensorflow warning: W tensorflow/core/data/root_dataset.cc:200] Optimization loop failed: CANCELLED: Operation was cancelled

I am getting this error when performing a custom training over a model, but the training seems to continue anyway... Is is something I should be worried about? I tried with several batch size and learning rates, and more hyperparameter combination…
SrTangente
  • 11
  • 2
0
votes
0 answers

How to train dialogflow to learn for license truck number

I've used dialog flow for conversation between our truck company and us by using dialog flow link with line app so when truck company inform us the chatbot will automatic reply to truck company and pass details to another group chat to our warehouse…
Sarayont Srisamudh
  • 1
0
votes
1 answer

How to work with Doc2Vec and which approach is better training the model on my dataset or using a pretrained model?

I am building a classification model for a dataset of items. Basically, I have 2 columns ex: Item name category unsalted butter dairy and eggs cheese dry grocery peanut butter cream dry grocery I did the required preprocessing to…
Mayar Alzerki
  • 23
  • 4
0
votes
0 answers

TensorFlow Custom Training Loop Returns None Gradients while using custom loss function

I am trying to perform Multi-Task Learning on the UTKFace Dataset. I can get the training done just fine. However, when I try to use graph execution using tf.function it doesn't work as the gradients returned is None. I am using…
Stevi G
  • 257
  • 1
  • 4
  • 13
1 2 3
4
5 6