Questions tagged [transfer-learning]

Transfer learning is the process of learning a set of characteristics from one data and applying this "knowledge" to another similar dataset (i.e. using the same model across datasets).

Filter by
Sorted by
Tagged with
0 votes
1 answer
60 views

What is the "Extract" token and how is the final Linear layer applied in GPT?

In the manuscript of GPT, the authors have given the following image: Questions: What is the final "Extract" (token?)? Is it the "END" token? How is the final linear layer ...
figs_and_nuts's user avatar
1 vote
1 answer
189 views

issue loading the ckpt file PytorchStreamReader failed reading zip archive: failed finding central directory

I am trying to load the ckpt file and getting error PytorchStreamReader failed reading zip archive: failed finding central directory Here is the code ...
Shruti's user avatar
  • 35
0 votes
0 answers
43 views

Different generated patches from original image using vision transformer (ViT)

I am using ViT for image classification, I scaled images in range of [-1,1], and I also padded images. Then, I used the following code to see the original image and generated patches, but the output ...
Zara Nz's user avatar
0 votes
0 answers
23 views

Why use sliding window input features in deep learning?

I was reading through the DNABERT paper and found that their input features were k-mers. This is equivalent to using rolling/sliding window features in the other common family of sequential problem, ...
Avatrin's user avatar
  • 131
0 votes
1 answer
66 views

Combine two separate models created via Transfer Learning?

Suppose have two 'image classification' models created by transfer learning on the same base model[1], each producing a different set of labels/classes. Trained at different times, with different ...
barryhunter's user avatar
0 votes
0 answers
13 views

How to analyze social media data to see its impact on a game's sales

I work for a console gaming giant. We forecasted the sales for a RPG game that was to be released few months back. But the actual sales was twice the forecast. This compelled the developers to ...
Ritik P. Nayak's user avatar
0 votes
0 answers
30 views

Prevent Overfitting in Transfer Learning with small data

I have built a feed forward neural network to predict heat pumps energy consumption. Now, i want to use this model as a domain for other heat pumps via transfer learning. I want to simulate the case ...
MBC_222's user avatar
0 votes
0 answers
24 views

Fine-tuning Pretrained Models for Web DOM Interaction Prediction Task

I am currently working on a side project that involves predicting changes to a webpage's DOM based on user interactions. The idea is to input the initial DOM state and a user interaction, and predict ...
DonnyDato's user avatar
0 votes
1 answer
36 views

Better results when adding a dropout layer before a single layer classifier - counter intuitive result

I am working on an multi-class image classification problem (with 9 classes), i am using a pretrained DenseNet121 (on ImageNet), i'm using Keras. i am using densenet as a feature extractor, with a ...
user062's user avatar
5 votes
3 answers
2k views

Further Training a pre-trained LLM

My goal is to use the general knowledge and language understanding of a pre-trained LLM and to continue training on a smaller domain specific corpus to improve the model's knowledge on the domain. ...
Arthuro's user avatar
  • 81
0 votes
0 answers
146 views

Fine-Tuning / Transfer learning results in worse performance

My task is creating a model for QA-purposes. I have only ~200 samples on a specific domain of questions. Using a pretrained like DeBERTa without any further changes results in f1 scores of ~35%. To ...
max245905's user avatar
0 votes
1 answer
27 views

Can I add a new output class to a decoder and train only the final layer?

I am wondering how to approach a project, where I would like to increase the number of output classes of an already trained network. I have very good reason to believe that the model has already ...
aqua's user avatar
  • 123
1 vote
1 answer
34 views

Speech to Text for Unsupported Language

I'm working on a project to plug the good old speech recognition in my app. However, I wish to do it in my country's dialect which is not supported by the major APIs like Azure, AWS, etc. My country's ...
GargantuanChad's user avatar
0 votes
0 answers
392 views

Model.train( ) returns automatically in YOLO v8

I'm trying to find the best sets of hyperparameters for my YOLO V8 model on my custom dataset with RayTune. I wanted to train the model with model.train() and return some of the evaluation metrics, ...
abdus samad's user avatar
0 votes
0 answers
41 views

The embedding output of bert

I want to get the embedding matrix of the Bert model (the input before the first block layers) to feed it into another architecture. I really appreciate it if you help me with that. Thanks
mansoor sh's user avatar
0 votes
1 answer
48 views

Single model or multiple models for predicting at each level in a multi-level classification problem

Given a flat structured data with features that can be considered hierarchical, where each feature is at a different level (e.g., Brand at the top level, Product, Color, and Size at different levels), ...
Kedharnath Kb's user avatar
0 votes
0 answers
8 views

Transfer learning applicability for Psychology experimental research

This is my first question so please be gentle! I am a Psychologist building a predictive model using experimental data and I want to know how I can do this using limited training data from part of my ...
Steve079150's user avatar
0 votes
0 answers
39 views

Training an Object Detection Model from scratch vs. pretrained weights

I have a question related to training a object detection model: Lets say I have trained a model for detecting 1 class with, say, 500 images including positive and negative samples and saved the best ...
Uce's user avatar
  • 1
0 votes
0 answers
19 views

Which CNN model to use for the classification(20 classes) of gemstones (diamonds, sapphire, ruby etc) based on digital photo images and huge data set?

Im trying to build CNN Model for the classification of precious stones (like diamonds, sapphire, ruby) based on digital images. So I have data set of labeled 150,000 gemstone certifications and the ...
Bob9710's user avatar
  • 101
0 votes
0 answers
50 views

LIME visualization for resnet50

I am implementing LIME on my resnet50 mode. There are 4 classes in the dataset. the code snippet of LIME: ...
Rezuana Haque's user avatar
0 votes
0 answers
36 views

Help to find strategy of transfer learning with a conditional pix2pix model

so I trained a similar model to pix2pix GAN to generate output images. I condition my model on three different types: there are 120 input images and 2 parameters (param_1, param_2). For each parameter,...
Rima's user avatar
  • 1
0 votes
0 answers
44 views

How many data points do we need for Transfer Learning for multi-class approach using Hugging Face models?

I'm just trying to see a few checkpoints on how many data points do need to do, let's say, a text classification transfer learning approach. Using the Bert-base model for example, and freezing all ...
Chronicles's user avatar
0 votes
0 answers
60 views

Make spacy NER model more robust to handle odd product code entity extraction

I am developing a NER model to extract product codes that are all over the place in terms of format and naming convention (AXEWAL719XA, AX-P20XXT-001, etc.). I started with the basic blank spacy('en') ...
scarpacci's user avatar
  • 101
0 votes
0 answers
19 views

Transfer learning from a single label task to a multi label classification task

I was thinking of using the weights of inception3 ...
Claudiu Creanga's user avatar
0 votes
0 answers
25 views

Transfer learning (or fine-tuning) pre-trained model on non-text data (PyTorch)

I am currently fine-tuning a sentiment analysis bert-based model using PyTorch Trainer from hugging face. So far, so good. I have easily managed to fine-tune the model on my text data. However, I'd ...
corvusMidnight's user avatar
0 votes
0 answers
32 views

How is model training affected after randomizing the weights of an intermediate layer of a pre-trained model?

Assuming that I have a deep learning model (let's say a ResNet) pretrained on a given dataset (let's say it is ImageNet). I load that model and randomize the weights of one of the intermediate layers, ...
Jefferson White's user avatar
0 votes
1 answer
77 views

Not able to understand Transfer Learning with Vgg16

So, I have to work with Vgg16 in my semester group project, and was following this to do transfer learning. I don't understand CNN much, but am learning currently. The very first problem was that ...
royal_awake's user avatar
1 vote
1 answer
86 views

Training a CNN in production on new data

How should I approach training a convolutional neural network in production on new data when I detect model performance degradation due to data or concept drift? Resources like this one and this one ...
Fijoy Vadakkumpadan's user avatar
1 vote
1 answer
266 views

CNN good results on train and test, bad results on real world data

I'm trying to build a neural network for an age detection task. Here some details : Dataset: I am using the "facial age" Kaggle dataset and the "UTKFace" dataset for a total of ...
Daniel_Fortesque's user avatar
0 votes
0 answers
971 views

Invalid argument error: logits and labels must be broadcastable

Whenever I try to execute this vgg16 code I get an error like this: ...
Pratik Pradhan's user avatar
0 votes
1 answer
250 views

How to increase FER2013 dataset validation_accuracy for only 3 classes i.e, happy,sad,neutral?

I am building a face emotion detection model using vgg16. Using FER2013 dataset for 7 classes i am getting= train_accuracy=97%, validation_accuracy=90%. but when i tried with 3 classes i.e, happy,sad,...
Pratik Pradhan's user avatar
0 votes
1 answer
49 views

Hello guys, is dimension reduction required for tensorflow? [closed]

I am working on face emotion detection using FER2013 dataset using tensorflow and vgg16 model. I am applying t-sne to my training dataset for dimensionality reduction. My question is that "is ...
Pratik Pradhan's user avatar
0 votes
2 answers
365 views

Autoencoder vs Pre-trained network for feature extraction

I wanted to know if anyone has any sort of guidance on what is better for image classification on a lot of classes (about 400) with a small amount of samples per class (around 20), for relatively big ...
MrStealYourFrog's user avatar
0 votes
1 answer
175 views

Weird consequence of not freezing layers in Neural Network

I was researching about "why are we freezing layers" and I came across the answer says "to not lose the information of pre-trained model" But; we are just freezing early layers (I ...
canP's user avatar
  • 91
0 votes
0 answers
39 views

Transfer learning: As simple as running trained models on new data?

So there's a domain of interest where the machine learning models are all specific to one entity. Let's call it a building. So there's a model made for every building. The literature in the domain all ...
There's user avatar
  • 121
1 vote
1 answer
61 views

How to summarize very large neural networks?

I am doing a lot of work with transfer learning at the moment (using keras and tensorflow if that is relevant). I am having a lot of issues in sufficiently summarizing the very large models. This post:...
Oskar's user avatar
  • 11
1 vote
0 answers
13 views

Dealing with little available data: transfer learning

Suppose I seek to predict a certain numerical value, whereby the data set which contains the predetermined correct labels is only very small. However, I'm also provided a large data set with a label ...
Richard's user avatar
  • 111
1 vote
1 answer
190 views

Pretrained vs. finetuned model

I have a doubt regarding terminology. When dealing with huggingface transformer models, I often read about "using pretrained models for classification" vs. "fine-tuning a pretrained ...
lazarea's user avatar
  • 289
1 vote
1 answer
342 views

Cannot achieve good result while Transfer Learning CIFAR-10 on ResNet50 - Keras

I'm trying to Transfer Learn ResNet50 for image classification of the CIFAR-10 dataset. It's stated in the original paper and also ResNet50 documentation on keras.io that the ResNet should have a ...
Mahdi's user avatar
  • 11
1 vote
2 answers
2k views

model.fit vs model.evaluate gives different results?

The following is a small snippet of the code, but I'm trying to understand the results of model.fit with train and test dataset vs the model.evaluate results. I'm not sure if they do not match up or ...
ptn77's user avatar
  • 111
0 votes
1 answer
4k views

No module named 'model'

I am trying to use the CoAtNet class in the following link CoAtNet Class from Github but I always have error while I am running the following lines: ...
Beba.S's user avatar
  • 37
0 votes
0 answers
52 views

Using Validation Set in Transfer Learning for Feature Extractor Preprocessor

I have a set of images of products. I am using transfer learning for images feature extraction in this way : I load a model (res-net, vgg) I add 2 dense layers, first one will be my features and the ...
W.314's user avatar
  • 101
-1 votes
3 answers
2k views

Transfer learning on YOLOv5 for character and shape detection

The task is to detect rotated alphanumeric characters embedded on colored shapes. We will have an aerial view of the object (from a UAS: Unarmed Aerial System), something of this sort: (One Uppercase ...
satan 29's user avatar
  • 103
0 votes
1 answer
3k views

How to freeze certain layers in models obtained from keras.applications

I am currrently trainning to use transfer learning on ResNet152 obtained from Keras Applications: ...
AAA's user avatar
  • 7
2 votes
0 answers
38 views

Transfer learning + Selective Fine Tuning

This is a two part question: Background: I have a rather small set (thousands) of medical images. I am using pretrained models off of kersa.applications without ...
Ali's user avatar
  • 121
1 vote
2 answers
423 views

Distribution Shift vs Transfer Learning

Transfer learning (TL) is a research problem in machine learning (ML) that focuses on storing knowledge gained while solving one problem and applying it to a different but related problem [1] ...
Carlos Mougan's user avatar
0 votes
1 answer
560 views

How does T5 model work on input and target data while transfer learning?

I am working on a project where I want the model to generate job description based on Role, Industry, Skills. I have trained my data and got the resultant output. I ...
10sha25's user avatar
  • 43
1 vote
1 answer
7k views

What is the difference between batch_encode_plus() and encode_plus()

I am doing a project using T5 Transformer. I have read documentations related to T5 Transformer model. While using T5Tokenizer I am kind of confused with tokenizing my sentences. Can someone please ...
10sha25's user avatar
  • 43
0 votes
1 answer
392 views

How many layers of a pretrained model shoud be frozen?

I am following a transfer learning example where the blogger has frozen the first 20 layers of MobileNet. My question is, is there any rule of thumb for how many layers should be frozen? What is the ...
imtiaz ul Hassan's user avatar
1 vote
1 answer
753 views

Is it possible to fine-tuning BERT by training it on multiple datasets? (Each dataset having it's own purpose)

BERT can be fine-tuned on a dataset for a specific task. Is it possible to fine-tune it on all these datasets for different tasks and then be utilized for these tasks instead of fine-tuning a BERT ...
Tony Jesuthasan's user avatar