56,048 questions
0
votes
0
answers
15
views
Node2vec embeddings for Fraud Detection?
I tried to improve XGBoost by stacking node2vec embeddings to the original features for fraud detection. In literature and papers this is frequently discussed as a good method to enhance prediction ...
- 171
0
votes
0
answers
17
views
Trained model has "training_args.bin" along with other essentials but I don't see "pytorch_model.bin"
I am training a model, and the trained model has "training_args.bin" along with other essentials. But I don't see "pytorch_model.bin". I read a bit and GPT'd, it says "...
-2
votes
0
answers
21
views
Handling NaN Values and Feature Consistency in Anomaly Detection with Isolation Forest [closed]
I am building a system for detecting anomalous player behavior. I have collected datasets from different games, selected the necessary features, and performed aggregation of these data using the ...
- 1
-2
votes
0
answers
35
views
For training a neural network, if i have label encoded my feature, is there need to scale it or normalize it?
I am working on a project which predict customer satisfaction score. I have several categorical features. One feature has 3 unique value while some has 59 and 1600 unique values. My question is can I ...
-1
votes
0
answers
23
views
Managing FastAPI + MLFlow dependencies
I've recently built an MVP for a dynamic FastAPI deployment template to reduce the time it takes for developers to deploy APIs for a given model. While working on the first new case, I've encountered ...
- 404
-4
votes
0
answers
72
views
How to improve handwriting recognition in an image [closed]
I'm trying to develop a system to read the handwriting in a chart within a written page, using a multimodal LLM. I'm using Google apps script So far I've experimented:
function openRouterApiRequest() {...
- 36.6k
-2
votes
0
answers
26
views
Streamlit Crashes on Image Upload [closed]
"OSError: cannot identify image file <_io.BytesIO object at 0x7fce02c76b80>"
I'm trying to deploy a ML model on Streamlit and I encountered this error as displayed above while trying ...
0
votes
0
answers
33
views
UNet is loading 1 epoch per day [closed]
I am not allowed to share my full code but I can share parts.
I have built a UNet that uses cache and shuffle buffers. It was working well and training fast until I realised any updates I made were ...
-3
votes
0
answers
41
views
How to upload a machine learning model (.pkl) to Hugging Face? [closed]
I'm new to machine learning and I’m trying to upload a simple model (.pkl file) to Hugging Face.
I’ve already created a model repository on Hugging Face, but I'm confused about how to upload the .pkl ...
- 1
0
votes
1
answer
27
views
How can I optimize a Streamlit dashboard for large CSV files to improve load time?
My Streamlit app reads and visualizes data from a large CSV file (~100MB). The app takes several seconds to load.
What are the best practices for improving performance��such as using @st.cache_data, ...
0
votes
0
answers
19
views
How to Extract Text Tables Images from PDFs while maintaining the structures
from unstructured library opensource one when i tried a pdf that have background images design patterns and XObjects in it this library also consider those as a images and store the path. so how can ...
- 11
0
votes
0
answers
24
views
How to covert pytorch to rknn?
I have exported the yolov5s PyTorch model from ultralytics/yolov5 and converted it to ONNX, using Google Colab, so that I can convert it to rknn in my rk3588 Radxa Rock 5c module.
%cd /content/
!pip ...
-1
votes
1
answer
41
views
Low Accuracy from Geospatial Random forest ML modeling problem - Training Exported from qGIS, SCP [closed]
I am doing a geospatial assessment integrated with ML modeling. The problem is the very low accuracy percentage, as more training features increases, it gets lower. What could be he solution to such a ...
- 15
0
votes
1
answer
44
views
How to preprocess date in Isolation Forest sklearn [closed]
I am using sklearn's IsolationForest model to detect anomalies on a time-series dataset. One of the features is date with the format MM-YYYY, the other features are numeric values.
What is the best ...
- 21
-1
votes
0
answers
44
views
How to use Stable Diffusion WebUI with ControlNet for pose-to-image generation (img2img)? [closed]
I'm trying to generate images using Stable Diffusion WebUI with ControlNet (pose) via img2img mode, and I want the output image to follow the pose of a reference skeleton image.
Here��s what I have so ...
- 15