Ah! My group uses machine learning in lots of ways in astronomy and while it’s great at what it does I absolutely relate to your scepticism :) I could easily foresee myself being put into an AI-deniers list despite using Mask-RCNN, pix2pix, Efficientnet, etc…
August 14, 2024 at 8:23 AM
Ah! My group uses machine learning in lots of ways in astronomy and while it’s great at what it does I absolutely relate to your scepticism :) I could easily foresee myself being put into an AI-deniers list despite using Mask-RCNN, pix2pix, Efficientnet, etc…
Master diabetic retinopathy detection! Optimize your model using CLAHE preprocessing, Test-Time Augmentation (TTA), and powerful ensembles like ResNet + EfficientNet to boost diagnostic accuracy. #DeepLearning #MedTech #Python #ComputerVision #AI https://ai.dappcrypto.org/r/m
July 1, 2026 at 1:52 PM
Master diabetic retinopathy detection! Optimize your model using CLAHE preprocessing, Test-Time Augmentation (TTA), and powerful ensembles like ResNet + EfficientNet to boost diagnostic accuracy. #DeepLearning #MedTech #Python #ComputerVision #AI https://ai.dappcrypto.org/r/m
www.nature.com/articles/s42...
𝗺𝘆 𝗺𝗼𝗺🪔
@bsky.app
🌐 #BOSCh
#DATA #bigdata #webinar
#CyberSecurity #business
#Crypto #Technology
#bitcoin #Digital #teCh #genAI
#ai #Medical #frenchTECh
#X #Linkedin #uk #HealthTech #DiGiTALhealth #5g #USA
#Quantum #Health #web3
#Cloud #MachineLearning
#SOciaLmeDia
𝗺𝘆 𝗺𝗼𝗺🪔
@bsky.app
🌐 #BOSCh
#DATA #bigdata #webinar
#CyberSecurity #business
#Crypto #Technology
#bitcoin #Digital #teCh #genAI
#ai #Medical #frenchTECh
#X #Linkedin #uk #HealthTech #DiGiTALhealth #5g #USA
#Quantum #Health #web3
#Cloud #MachineLearning
#SOciaLmeDia
A composite scaling network of EfficientNet for improving spatial domain identification performance - Communications Biology
EfNST improves the identification of spatial domains using EfficientNet for multi-scale image features. It effectively reveals tissue structures and marker genes, aiding in the understanding of spatia...
www.nature.com
November 26, 2024 at 4:28 AM
And these models are just classifiers (ResNet / EfficientNet) and importantly are NOT Generative Adversarial Networks, which would have a component that would generate imagery for obvious reasons. Some approaches train classifiers on commercial adult pornography for human review
February 11, 2025 at 6:43 PM
And these models are just classifiers (ResNet / EfficientNet) and importantly are NOT Generative Adversarial Networks, which would have a component that would generate imagery for obvious reasons. Some approaches train classifiers on commercial adult pornography for human review
I used EfficientNet (lmz/candle-efficientnet from huggingface), there's also an option to use convnext but it runs like unusably slow for me because i can't use gpu accel
May 24, 2026 at 9:21 PM
I used EfficientNet (lmz/candle-efficientnet from huggingface), there's also an option to use convnext but it runs like unusably slow for me because i can't use gpu accel
Как оптимизировать модель детекции диабетической ретинопатии? Используйте CLAHE, TTA и ансамбли ResNet/EfficientNet. Это поможет избежать ошибок классификации даже при высокой уверенности модели. #AI #DeepLearning #Python #MedicalAI #ComputerVision https://ai.dappcrypto.org/r/m
July 1, 2026 at 1:58 PM
Как оптимизировать модель детекции диабетической ретинопатии? Используйте CLAHE, TTA и ансамбли ResNet/EfficientNet. Это поможет избежать ошибок классификации даже при высокой уверенности модели. #AI #DeepLearning #Python #MedicalAI #ComputerVision https://ai.dappcrypto.org/r/m
Researchers at Vel Tech Institute trained deep learning models on cough sounds to detect COVID-19.
Their ResNet-18 model achieved 98.5% accuracy and only 1% false positives, showing strong potential for rapid, low-cost screening.
www.nature.com/articles/s41...
Their ResNet-18 model achieved 98.5% accuracy and only 1% false positives, showing strong potential for rapid, low-cost screening.
www.nature.com/articles/s41...
ResNet-EfficientNet powered framework for high-precision cough-based classification of infectious diseases - Scientific Reports
Scientific Reports - ResNet-EfficientNet powered framework for high-precision cough-based classification of infectious diseases
www.nature.com
November 6, 2025 at 6:32 PM
Researchers at Vel Tech Institute trained deep learning models on cough sounds to detect COVID-19.
Their ResNet-18 model achieved 98.5% accuracy and only 1% false positives, showing strong potential for rapid, low-cost screening.
www.nature.com/articles/s41...
Their ResNet-18 model achieved 98.5% accuracy and only 1% false positives, showing strong potential for rapid, low-cost screening.
www.nature.com/articles/s41...
Hi. Firstly, Sethi's 2020 work is good, but I would NOT recommend "VGGish" any more - it's vastly outdated and modern CNNs do this job much better. A very common choice these days is EfficientNet, and lots of pretrained audio models are based on it.
April 10, 2025 at 9:47 AM
Hi. Firstly, Sethi's 2020 work is good, but I would NOT recommend "VGGish" any more - it's vastly outdated and modern CNNs do this job much better. A very common choice these days is EfficientNet, and lots of pretrained audio models are based on it.
Update on tagging model:
After the test run, I used a bigger model (Efficientnet V2 S) as the base and then also trained for more than one epoch... now two epochs!
This is actually pretty good labeling 👀👀👀
(#nsfw)
After the test run, I used a bigger model (Efficientnet V2 S) as the base and then also trained for more than one epoch... now two epochs!
This is actually pretty good labeling 👀👀👀
(#nsfw)
August 11, 2023 at 11:42 PM
Update on tagging model:
After the test run, I used a bigger model (Efficientnet V2 S) as the base and then also trained for more than one epoch... now two epochs!
This is actually pretty good labeling 👀👀👀
(#nsfw)
After the test run, I used a bigger model (Efficientnet V2 S) as the base and then also trained for more than one epoch... now two epochs!
This is actually pretty good labeling 👀👀👀
(#nsfw)
Glad you're finding it useful! We've also done building height from single shot images. findingspress.org/article/1166...
We got an accuracy of 1.21 m - so it works pretty well =) @alexo.bsky.social
We got an accuracy of 1.21 m - so it works pretty well =) @alexo.bsky.social
Single-Image Building Height Estimation Using EfficientNet: A Simplified, Scalable Approach | Published in Findings
By Alexander W Olson, Shoshanna Saxe. This study introduces an effective method using EfficientNet for single-image building height estimation, demonstrating high accuracy and simplicity compared to t...
findingspress.org
November 20, 2024 at 12:12 AM
Glad you're finding it useful! We've also done building height from single shot images. findingspress.org/article/1166...
We got an accuracy of 1.21 m - so it works pretty well =) @alexo.bsky.social
We got an accuracy of 1.21 m - so it works pretty well =) @alexo.bsky.social
New at Findings Press: , fediscience.org/@Findingspre..., Single-Image Building Height Estimation Using EfficientNet: A Simplified, Scalable Approach findingspress.org/article/1166...
Findings Press (@Findingspress@fediscience.org)
Single-Image Building Height Estimation Using EfficientNet: A Simplified, Scalable Approach https://findingspress.org/article/116609-single-image-building-height-estimation-using-efficientnet-a-simpli...
fediscience.org
May 10, 2024 at 10:55 PM
New at Findings Press: , fediscience.org/@Findingspre..., Single-Image Building Height Estimation Using EfficientNet: A Simplified, Scalable Approach findingspress.org/article/1166...
New publication from our team🎉 besjournals.onlinelibrary.wiley.com/doi/pdfdirec...
We benchmarked citizen science vs. EfficientNet & DeepFaune on 51,588 camera-trap images. Citizen science had high precision but variable recall; AI improved recall for several species, yet both were weaker at night.
We benchmarked citizen science vs. EfficientNet & DeepFaune on 51,588 camera-trap images. Citizen science had high precision but variable recall; AI improved recall for several species, yet both were weaker at night.
besjournals.onlinelibrary.wiley.com
October 8, 2025 at 12:16 PM
New publication from our team🎉 besjournals.onlinelibrary.wiley.com/doi/pdfdirec...
We benchmarked citizen science vs. EfficientNet & DeepFaune on 51,588 camera-trap images. Citizen science had high precision but variable recall; AI improved recall for several species, yet both were weaker at night.
We benchmarked citizen science vs. EfficientNet & DeepFaune on 51,588 camera-trap images. Citizen science had high precision but variable recall; AI improved recall for several species, yet both were weaker at night.
🚀 Advancing Brain MRI Image Classification with Deep Learning 🧠
I’ve recently completed a model focused on improving brain MRI classification using EfficientNet and Grad-CAM. If you’re into AI and Healthcare, check it out!
medium.com/@robinsonjas...
#AI #DeepLearning #MedicalImaging #ExplainableAI
I’ve recently completed a model focused on improving brain MRI classification using EfficientNet and Grad-CAM. If you’re into AI and Healthcare, check it out!
medium.com/@robinsonjas...
#AI #DeepLearning #MedicalImaging #ExplainableAI
Advancing Brain MRI Image Classification with Deep Learning
AI-powered medical imaging-based classification model with Grad-CAM for explainability.
medium.com
November 17, 2024 at 6:22 PM
🚀 Advancing Brain MRI Image Classification with Deep Learning 🧠
I’ve recently completed a model focused on improving brain MRI classification using EfficientNet and Grad-CAM. If you’re into AI and Healthcare, check it out!
medium.com/@robinsonjas...
#AI #DeepLearning #MedicalImaging #ExplainableAI
I’ve recently completed a model focused on improving brain MRI classification using EfficientNet and Grad-CAM. If you’re into AI and Healthcare, check it out!
medium.com/@robinsonjas...
#AI #DeepLearning #MedicalImaging #ExplainableAI
Mahmudul Hasan
Deep Learning for Breast Cancer Detection: Comparative Analysis of ConvNeXT and EfficientNet
https://arxiv.org/abs/2505.18725
Deep Learning for Breast Cancer Detection: Comparative Analysis of ConvNeXT and EfficientNet
https://arxiv.org/abs/2505.18725
May 27, 2025 at 5:17 PM
Mahmudul Hasan
Deep Learning for Breast Cancer Detection: Comparative Analysis of ConvNeXT and EfficientNet
https://arxiv.org/abs/2505.18725
Deep Learning for Breast Cancer Detection: Comparative Analysis of ConvNeXT and EfficientNet
https://arxiv.org/abs/2505.18725
Leveraging Vision-Language Models for Resource Constrained Settings
Anna Bair, Madan Ravi Ganesh, Devin Willmott, J Zico Kolter
Action editor: Massimiliano Mancini
https://openreview.net/forum?id=cYOKSg60jC
#efficientnet #leveraging #trained
Anna Bair, Madan Ravi Ganesh, Devin Willmott, J Zico Kolter
Action editor: Massimiliano Mancini
https://openreview.net/forum?id=cYOKSg60jC
#efficientnet #leveraging #trained
August 10, 2026 at 12:20 PM
Leveraging Vision-Language Models for Resource Constrained Settings
Anna Bair, Madan Ravi Ganesh, Devin Willmott, J Zico Kolter
Action editor: Massimiliano Mancini
https://openreview.net/forum?id=cYOKSg60jC
#efficientnet #leveraging #trained
Anna Bair, Madan Ravi Ganesh, Devin Willmott, J Zico Kolter
Action editor: Massimiliano Mancini
https://openreview.net/forum?id=cYOKSg60jC
#efficientnet #leveraging #trained
EfficientNet (2019) asked a simple question: how do you scale neural networks optimally? Width? Depth? Resolution? Answer: all three, carefully balanced. Result? Better accuracy with 10x fewer parameters. Compound scaling became the blueprint for efficient model design.
November 10, 2025 at 2:01 PM
EfficientNet (2019) asked a simple question: how do you scale neural networks optimally? Width? Depth? Resolution? Answer: all three, carefully balanced. Result? Better accuracy with 10x fewer parameters. Compound scaling became the blueprint for efficient model design.
Equating the architecture and dataset, but switching objective from glimpse prediction to caption embedding (MPNet; sGSN) or multi-class object prediction (cGSN), reduces the alignment. Furthermore, no related/SOTA model (36 tested; Table S5) outperforms GPN-R-SimCLR => a new SOTA model! 9/14
November 18, 2025 at 12:37 PM
Equating the architecture and dataset, but switching objective from glimpse prediction to caption embedding (MPNet; sGSN) or multi-class object prediction (cGSN), reduces the alignment. Furthermore, no related/SOTA model (36 tested; Table S5) outperforms GPN-R-SimCLR => a new SOTA model! 9/14
We present AnomalyMatch, an anomaly detection framework combining the semi-supervised FixMatch algorithm using EfficientNet classifiers with active learning. By treating anomaly detection as a semi-supervised binary classification problem, we [2/8 of https://arxiv.org/abs/2505.03509v1]
May 7, 2025 at 6:05 AM
We present AnomalyMatch, an anomaly detection framework combining the semi-supervised FixMatch algorithm using EfficientNet classifiers with active learning. By treating anomaly detection as a semi-supervised binary classification problem, we [2/8 of https://arxiv.org/abs/2505.03509v1]
Siberian radioheliograph image classification using ensemble of CLIP, EfficientNet and CatBoost models. Yaroslav Egorov https://arxiv.org/abs/2507.04211
July 8, 2025 at 5:04 AM
Siberian radioheliograph image classification using ensemble of CLIP, EfficientNet and CatBoost models. Yaroslav Egorov https://arxiv.org/abs/2507.04211
Published in #RASTI RAS Techniques & Instruments: "AnomalyMatch: discovering rare objects of interest with semisupervised and active learning", Gómez et al. This is Fig. 1: please visit academic.oup.com/rasti/articl... to read the paper. @royalastrosoc.bsky.social
July 17, 2026 at 8:34 AM
Published in #RASTI RAS Techniques & Instruments: "AnomalyMatch: discovering rare objects of interest with semisupervised and active learning", Gómez et al. This is Fig. 1: please visit academic.oup.com/rasti/articl... to read the paper. @royalastrosoc.bsky.social
今日のAIニュース
画像特徴量抽出モデルの比較：ResNetの進化と他のモデルとの違い | Zennの「機械学習」のフィード
この記事は、画像特徴量抽出に用いられる主要なモデル（ResNet、EfficientNet、VGG、Inception）の比較解説です。
ResNetは層の深さによる種類があり、EfficientNetは高効率、VGGはシンプルだが高負荷、Inceptionは多スケール特徴抽出に優れます。
用途に応じてモデルを選択する必要があり、特にResNet50またはEfficientNet-B3が多くのタスクでバランス良く推奨されています。
画像特徴量抽出モデルの比較：ResNetの進化と他のモデルとの違い | Zennの「機械学習」のフィード
この記事は、画像特徴量抽出に用いられる主要なモデル（ResNet、EfficientNet、VGG、Inception）の比較解説です。
ResNetは層の深さによる種類があり、EfficientNetは高効率、VGGはシンプルだが高負荷、Inceptionは多スケール特徴抽出に優れます。
用途に応じてモデルを選択する必要があり、特にResNet50またはEfficientNet-B3が多くのタスクでバランス良く推奨されています。
画像特徴量抽出モデルの比較：ResNetの進化と他のモデルとの違い
🟢 はじめに画像の特徴量を抽出する際に、最もよく使われるのが**CNN（畳み込みニューラルネットワーク）**です。その中でも ResNet（Residual Networks） は、さまざまなバージョンがあり、用途に応じて選択されます。また、ResNet以外にも、EfficientNet・VGG・Inception・MobileNet など、画像の特徴量を抽出するための強力なモデルが存在します。今
zenn.dev
February 27, 2025 at 7:13 AM
今日のAIニュース
画像特徴量抽出モデルの比較：ResNetの進化と他のモデルとの違い | Zennの「機械学習」のフィード
この記事は、画像特徴量抽出に用いられる主要なモデル（ResNet、EfficientNet、VGG、Inception）の比較解説です。
ResNetは層の深さによる種類があり、EfficientNetは高効率、VGGはシンプルだが高負荷、Inceptionは多スケール特徴抽出に優れます。
用途に応じてモデルを選択する必要があり、特にResNet50またはEfficientNet-B3が多くのタスクでバランス良く推奨されています。
画像特徴量抽出モデルの比較：ResNetの進化と他のモデルとの違い | Zennの「機械学習」のフィード
この記事は、画像特徴量抽出に用いられる主要なモデル（ResNet、EfficientNet、VGG、Inception）の比較解説です。
ResNetは層の深さによる種類があり、EfficientNetは高効率、VGGはシンプルだが高負荷、Inceptionは多スケール特徴抽出に優れます。
用途に応じてモデルを選択する必要があり、特にResNet50またはEfficientNet-B3が多くのタスクでバランス良く推奨されています。