27,175 questions
- Bountied 0
- Unanswered
- Frequent
- Score
- Trending
- Week
- Month
- Unanswered (my tags)
Score of -1
0 answers
35 views
How to use `class_weights` for the validation data when fitting a tensorflow model
I am training a tensorflow model and want to use class_weight for evaluating both the training batches and the validation batches. However, I am only able to use the weights with the training data. ...
- reputation score 44
Advice
0
votes
0
replies
56
views
Spent months building optimizers/CNNs from scratch in NumPy/CuPy — not sure what to build next, would appreciate direction
I have been teaching myself ML by building everything from raw math no heavy libraries like PyTorch, just NumPy/CuPy and derivatives worked out by hand. Wanted to share where I've landed and get some ...
- reputation score 1
Best practices
0
votes
0
replies
51
views
what are minimum image per class requied for max efficiency for efficientnet-lite0 and mobilenetv2 models
I am working on an image classification project for plant disease detection using transfer learning models such as MobileNetV2 and EfficientNetLite. My datasets include 11 classes for maize diseases ...
- reputation score 1
Score of 1
1 answer
85 views
CUDAExecutionProvider fails to initialize, InsightFace uses CPU only
I'm trying to run InsightFace with GPU acceleration on a Windows 11 machine with an RTX 4060 Laptop GPU. However, InsightFace is unable to use CUDA and always falls back to CPUExecutionProvider.
...
- reputation score 19
Advice
0
votes
1
replies
74
views
Trigram Language Model :Two implementations give different loss, are they equivalent?
I am implementing a trigram character-level language model following Andrej Karpathy's makemore series. I have two implementations and I want to understand if they are mathematically equivalent or ...
- reputation score 1
Advice
1
vote
2
replies
101
views
Will a 80 GB GPU and a 48 GB GPU give identical results on an open source text-to-video model for the same quantization and seed?
I am considering to buy GPUs for my project of open source text-to-video models like ltx-2-19b (lightricks) or wan-v2.2-a14b. I read online that the same configuration/quantization and seed will give ...
- reputation score 49
Best practices
2
votes
0
replies
58
views
What is the practical approach to become an AI Developer to Build AI powered apps/products?
I request Senior AI Developers to give answer take this question very seriously as most of the students trying to embark in AI your time and advice is very valuable for us so I request all the AI ...
- reputation score 1
Score of 3
1 answer
144 views
Problem in tensor shape difference between training and and testing data when running a CNN model
I have a problem when running the saved model of a CNN. When I train the model and the input tensor size of the first linear layer is set to the shape for the training data after it is flattened and ...
- reputation score 21
Best practices
0
votes
0
replies
37
views
Tuning for SCAFFOLD Server + Client Learning Rates η_g and η_i?
for my project on training a binary classifier on a residual CNN - data dimension (12, 4096), I am simulating federated training using FedAvg, FedProx and SCAFFOLD as the aggregation strategies.
The ...
- reputation score 75
Advice
0
votes
2
replies
52
views
How do we deal with partially labelled data in Classification problems
How do we deal with ML situations where part of the data has labels or ground truths and the rest doesn't?
- reputation score 391
Best practices
0
votes
0
replies
69
views
Best practices for image annotation consistency in object detection projects
I am working on a computer vision dataset preparation workflow for object detection training.
The project involves image annotation using bounding boxes and polygon segmentation for AI model training.
...
- reputation score 1
Advice
1
vote
1
replies
136
views
How to Evaluate my RAG System using Local / free tier API model
I have build a self-evaluating RAG System now i want to evaluate it so i can know what my system performance for that i used DeeEval framework but the thing is i don’t have OpenAI API key as i’m ...
- reputation score 1
Score of 0
1 answer
74 views
Issue when loading model with hub.KerasLayer layer... NameError: Exception encountered when calling Lambda.call()
I am doing some NLP exercises. I am unable to use a saved model with a hub.KerasLayer layer :(
Here are the steps I done:
Create the model - OK
Compile the model - OK
Fit/Train the model - OK
...
- reputation score 3
Advice
0
votes
0
replies
56
views
Regarding's MONAI WarmupCosineSchedule with AdamW, should scheduler.step() be called per batch, and does optimizer lr define the peak lr?
I am training a PyTorch segmentation model and using:
torch.optim.AdamW
monai.optimizers.WarmupCosineSchedule
My optimizer:
optimizer = torch.optim.AdamW(
model.parameters(),
lr=1e-4,
...
- reputation score 163
Tooling
0
votes
0
replies
86
views
What are the best methods or libraries to extract Arabic text from images or PDFs without using Surya OCR?
I am working on a project where I need to extract Arabic text from images and PDF documents. Currently, I am using Surya OCR, but I would like to explore other alternatives that might provide better ...
- reputation score 1