How to explain the #output of your #MachineLearning model? 🤔

📊 In this tutorial we'll explore how to use #SHAP values to explain and improve #ML models, delving deeper into specific use cases.

📚 Full tutorial: https://www.giskard.ai/knowledge/opening-the-black-box-using-shap-values-to-explain-and-enhance-machine-learning-models

Machine learning models are powerful but hard to interpret. However, SHAP values can help you understand how model features impact predictions.

https://www.datacamp.com/tutorial/introduction-to-shap-values-machine-learning-interpretability

#shap #machinelearning #XAI

SHAP を用いた機械学習への解釈性付与│データ分析とケモインフォ
https://datadriven-rnd.com/shap/#:~:text=SHAP%E3%81%A8%E3%81%AF,%E3%82%92%E6%B1%82%E3%82%81%E3%82%8B%E3%81%93%E3%81%A8%E3%81%8C%E3%81%A7%E3%81%8D%E3%82%8B%E3%80%82

#datascience #ai #explainableai #shap

'Faith-Shap: The Faithful Shapley Interaction Index', by Che-Ping Tsai, Chih-Kuan Yeh, Pradeep Ravikumar.

http://jmlr.org/papers/v24/22-0202.html

#shapley #shap #games

'On the Complexity of SHAP-Score-Based Explanations: Tractability via Knowledge Compilation and Non-Approximability Results', by Marcelo Arenas, Pablo Barcelo, Leopoldo Bertossi, Mikael Monet.

http://jmlr.org/papers/v24/21-0389.html

#shap #shapley #boolean

Fab wee feature from @GrahamUney about why he fell for his ace little corner of Cumbria.

#Wildhaweswater
#Bampton
#Shap
#LakeDistrict
#Cumbria

https://www.ukhillwalking.com/articles/features/my_move_to_the_mountains_life_in_the_lake_district_east-15103

Outrage as hundreds of New Balance trainers found in skip

https://www.bbc.co.uk/news/uk-england-cumbria-64855222

Surely these could have been given to those who can't afford footwear, at the very least.

#Shap #Cumbria #SportsGear #ClosedStore #Waste #Environment #CostOfLivingCrisis

Attention #AI enthusiasts: Don't fall prey to the pitfalls of careless #XAI usage! Beware of unreliable feature importance values when using #SHAP with preprocessed features. Let's avoid wasting our hard work and strive for accurate and meaningful results. #MachineLearning

https://blog.esciencecenter.nl/the-hidden-dangers-of-using-xai-carelessly-88f1d5fc3432

Finally made it to the road in the middle of the #M6 near #Shap. I know this shouldn't excite me, but it does 😁

Finally made it to the road in the middle of the #M6 near #Shap. I know this shouldn't excite me, but it does 😁

there’s a lot of really cool stuff in #MLEngineering that amounts to “train another model”. like using #SHAP to automate feature selection (first you have to train a model though). or #ConceptSHAP where you train simple linear models on the output of each neural net layer. or anomaly detection, or autoencoders, or...

Any #XAI #ExplainableAI libraries that actually work with keras tabular models with "big" data??? I've tried #Shap and #alibi but couldn't get them to work.
#machinelearning #artificialintelligence #python

Ya está abierto el registro para nuestra reunión de enero: "🐝 Drones conectados y cómo interpretar el modelo", este mes en KSchool

https://www.meetup.com/pydata-madrid/events/290889363/

¡Nos vemos el jueves 19 a las 19:00! Y después, networking 🍕

#PyDataMadrid #PyData #drones #connectedtechnology #explainability #shap

Have you ever used SHAP with image data?

In this #Python tutorial, we explore how to:
- Fine-tune ResNet-18 using #PyTorch
- Evaluate the model using MSE and scatter plots
- Interpret the model using #DeepSHAP
- Correct the model through better data collection
- Discuss further improvements using image augmentation

No paywall link: https://towardsdatascience.com/using-shap-to-debug-a-pytorch-image-regression-model-4b562ddef30d?source=friends_link&sk=7eb3016839186f1ba2a6f1f105f8ff64

#DataScience #MachineLearning #DeepLearning #SHAP #IML #XAI

Have you ever used SHAP with image data?

In this #Python tutorial, we explore how to:
- Fine-tune ResNet-18 using #PyTorch
- Evaluate the model using MSE and scatter plots
- Interpret the model using #DeepSHAP
- Correct the model through better data collection
- Discuss further improvements using image augmentation

No paywall link: https://towardsdatascience.com/using-shap-to-debug-a-pytorch-image-regression-model-4b562ddef30d?source=friends_link&sk=7eb3016839186f1ba2a6f1f105f8ff64

#DataScience #MachineLearning #DeepLearning #SHAP #IML #XAI

Have you ever used SHAP with image data?

In this Python tutorial, we explore how to:
- Fine-tune ResNet-18 using #PyTorch
- Evaluate the model using MSE and scatter plots
- Interpret the model using #DeepSHAP
- Correct the model through better data collection
- Discuss how image augmentation could further improve the model

#DataScience #MachineLearning #Python #SHAP #IML #XAI

No paywall link: https://towardsdatascience.com/using-shap-to-debug-a-pytorch-image-regression-model-4b562ddef30d?source=friends_link&sk=7eb3016839186f1ba2a6f1f105f8ff64

Working on a new article "Debugging a PyTorch Image Model with SHAP"

A mini-car takes images of a track as input. Using these, a model is used to make predictions which then turn the car.

Using SHAP, we can see the model uses background pixels when making predictions. This causes problems when we move the car to a new location.

Should be out next week :)

#DataScience #MachineLeaning #SHAP #IML #XAI #Interpretability

RT @Gabors_Data@twitter.com

Did you know we had a new release for @Gabors_Data@twitter.com case study repo? It's v0.8.3 "Ethics Gradient"
https://github.com/gabors-data-analysis/da_case_studies/releases/tag/v0.8.3
Mostly #Python updates. + adds #Shap. Enjoy. Comment.

🐦🔗: https://twitter.com/Gabors_Data/status/1608499200135921669

#DailyMegalith The Thunder Stone, an impressive glacial erratic in the #Shap landscape, #Cumbria, #England. (2014)
Maybe this stone was the inspiration for the other monuments - there are a number of standing stones and stone circles nearby. https://social.tchncs.de/media/Kvqluf6xygnMMkLWjKE