All Questions
13,343
questions
0
votes
0
answers
7
views
Unet from segmentation_models_pytorch stalling in training
I have been following a tutorial on training a segmentation model on a custom dataset, but it refuses to make any progrees in training the model.
This is my model setup
import ...
- 13
-3
votes
0
answers
25
views
How do I Install pytorch without GPU dependencies?
I have a project that depends on torch==2.0.1, but it will run on CPU not GPU. torch has some large cuda/cublas/cudnn dependencies that I believe are only needed when running on GPU. These packages, ...
- 1
0
votes
1
answer
24
views
load a model from checkpoint folder in pyTorch
I am trying to load a model from a certain checkpoint and use it for inference. The checkpoint folder looks like this. How do I load the model in torch from this folder. The resources I could find are ...
- 652
0
votes
0
answers
16
views
LSTM With DiffSharp
I'm trying to convert the full code of the example shown on this webpage: https://machinelearningmastery.com/lstm-for-time-series-prediction-in-pytorch/ which involves implementation of an LSTM.
My ...
- 171
0
votes
0
answers
25
views
RuntimeError: Given groups=1, expected weight to be at least 1 at dimension 0, but got weight of size [0, 16, 1, 1] instead in YOLOv8 Model Training
I am trying to train a YOLOv8 model using the yolov8n-pose.pt pretrained model with the following configuration in config.yaml:
# Data
path: C:\Users\Denis\OneDrive\Documents\Project\Workout Assistant\...
1
vote
0
answers
27
views
Unpickling Error: magic_number = pickle_module.load(f, **pickle_load_args) _pickle.UnpicklingError: invalid load key, 'v'
When I am trying to load a .pt file i am seeing the following error,
str1='Dataset/ALL_feats_cgqa.pt'
m = torch.load(str1)
the error is as follows,
File "/home/Storage1/pythonCodeArea/train.py&...
- 233
0
votes
0
answers
14
views
PyTorch zeroGrad() Equivalent in DiffSharp?
What is the PyTorch zeroGrad() Equivalent in DiffSharp? Doesn't seem to exist any quivalent member function of the Optimizer class.
- 171
1
vote
1
answer
32
views
"OSError: [WinError 127] The specified procedure could not be found" when importing torchtext
When importing get_tokenizer from torchtext:
from torchtext.data.utils import get_tokenizer
it throws this error:
Traceback (most recent call last):
File "D:\PythonProjects\FakeNews\train.py&...
0
votes
0
answers
23
views
Python multiprocessing demon CV Model dies
I am trying to write such app. Where I am need three demon Process: VideoReader, CV Model Logic, Data REST Sender and postprocessing.The model use PyTorch, so it is necessary to run in the main ...
- 75
0
votes
0
answers
12
views
How to Track Attention Weights of a Transformer Model with Comet?
I am working on a Transformer model for a translation task and want to track attention weights using Comet. My model consists of 2 layers with 2 attention heads each. I am interested in understanding ...
1
vote
0
answers
25
views
Converting a PyTorch ONNX model to TensorRT engine - Jetson Orin Nano
I'm trying to convert a ViT-B/32 Vision Transformer model from the UNICOM repository on a Jetson Orin Nano. The model's Vision Transformer class and source code is here.
I use the following code to ...
- 2,536
-1
votes
0
answers
25
views
Poor Performance and Signs of Overfitting When Fine-Tuning BART with Adapters on CNN/DailyMail Dataset
I am currently fine-tuning the BART model with adapters for a summarization task using the CNN/DailyMail dataset. I've noticed that the model shows poor performance and signs of overfitting. Below is ...
- 349
2
votes
3
answers
585
views
FBGEMM load error trying to use PyTorch on Windows
I'm working on a code that uses Whisper, and I need PyTorch with CUDA to improve the speed of the model execution, I have CUDA installed (verified using nvidia-smi command where it shows that I have ...
- 21
0
votes
1
answer
26
views
How to scan over a list of keywords in parquet using polars scan_parquet
I have a parquet file with metadata of categories. I want to scan this parquet file using scan_parquet from polars like below:
filtered_df = (
pl.scan_parquet(parquet_file)
...
- 13
-1
votes
0
answers
18
views
Multitasking bert for multilabel classification of 5 classes [duplicate]
I built 5 BioClinicalBERT-based models (finetuned bert) to predict labels for medical records for the following categories:
specialties = ["aud","den","oph","oto&...
