corpus.blog
41554 blogs · [ { "id": "01a08779-8ca3-71dd-8be7-38d779713659", "title": "Building Visual Neurons: image + video generation with Next.js, Replicate, and Amazon Bedrock", "url": "https://francescopochetti.com/building-visual-neurons-image-video-generation-with-next-js-replicate-and-amazon-bedrock/", "published_at": "2025-11-11T10:38:32+00:00" }, { "id": "01a08779-8ca4-7328-91af-0673336d838c", "title": "Byte Pair Encoding: building the GPT tokenizer with Karpathy", "url": "https://francescopochetti.com/byte-pair-encoding-building-the-gpt-tokenizer-with-karpathy/", "published_at": "2024-03-17T17:28:38+00:00" }, { "id": "01a08779-8ca4-7328-91af-067333b5a58a", "title": "CUDA for Python Programmers", "url": "https://francescopochetti.com/cuda-for-python-programmers/", "published_at": "2024-02-24T12:59:32+00:00" }, { "id": "01a08779-8ca4-7328-91af-067333c2eb83", "title": "LogoNet: the journey to an AWS-powered cloud application running ControlNet on SageMaker async endpoints", "url": "https://francescopochetti.com/logonet-the-journey-to-an-aws-powered-cloud-application-running-controlnet-on-sagemaker-async-endpoints/", "published_at": "2023-06-04T21:12:38+00:00" }, { "id": "01a08779-8ca4-7328-91af-067334003e83", "title": "A visual deep dive into the Transformer’s architecture: turning Karpathy’s masterclass into pictures", "url": "https://francescopochetti.com/a-visual-deep-dive-into-the-transformers-architecture-turning-karpathys-masterclass-into-pictures/", "published_at": "2023-03-09T16:25:47+00:00" }, { "id": "01a08779-8ca4-7328-91af-067334aebdce", "title": "Fastai Course Part 2 2022: Understanding CallBacks", "url": "https://francescopochetti.com/fastai-course-part-2-2022-understanding-callbacks/", "published_at": "2023-01-17T17:08:56+00:00" }, { "id": "01a08779-8ca4-7328-91af-067334b645b7", "title": "Amazon SageMaker Shadow Deployment: Semantic Segmentation from the HuggingFace Hub", "url": "https://francescopochetti.com/amazon-sagemaker-shadow-deployment-semantic-segmentation-from-the-huggingface-hub/", "published_at": "2022-12-27T21:37:23+00:00" }, { "id": "01a08779-8ca4-7328-91af-0673353f00c0", "title": "Benchmarking TorchVision ResNet18 on EC2 NVIDIA GPU with TensorRT and Amazon SageMaker Neo", "url": "https://francescopochetti.com/benchmarking-torchvision-resnet18-on-local-gpu-with-tensorrt-and-amazon-sagemaker-neo/", "published_at": "2022-11-30T06:39:13+00:00" }, { "id": "01a08779-8ca4-7328-91af-0673361f5aff", "title": "Benchmarking TorchVision ResNet18 on Amazon SageMaker CPU, GPU, and Inferentia instances (with a Neo twist)", "url": "https://francescopochetti.com/benchmarking-torchvision-resnet18-on-amazon-sagemaker-cpu-gpu-and-inferentia-instances-with-a-neo-twist/", "published_at": "2022-11-12T21:32:12+00:00" }, { "id": "01a08779-8ca4-7328-91af-067336a0caa3", "title": "Neural Magic: Training YoloV5 with Sparse Transfer Learning and deploying to Amazon SageMaker with a custom Docker container", "url": "https://francescopochetti.com/neural-magic-training-yolov5-with-sparse-transfer-learning-and-deploying-to-amazon-sagemaker-with-a-custom-docker-container/", "published_at": "2022-08-31T20:24:16+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a12054b9c", "title": "Training and Deploying a fully Dockerized License Plate Recognition app with IceVision, Amazon Textract and FastAPI", "url": "https://francescopochetti.com/training-and-deploying-a-fully-dockerized-license-plate-recognition-app-with-icevision-amazon-textract-and-fastapi/", "published_at": "2022-06-26T16:18:34+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a12b0bdd5", "title": "Blurry faces: Training, Optimizing and Deploying a segmentation model on Amazon SageMaker with NVIDIA TensorRT and NVIDIA Triton", "url": "https://francescopochetti.com/blurry-faces-a-journey-from-training-a-segmentation-model-to-deploying-tensorrt-to-nvidia-triton-on-amazon-sagemaker/", "published_at": "2022-05-22T17:23:38+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a137c2237", "title": "How much data should you allocate to training and validation?", "url": "https://francescopochetti.com/how-big-should-training-and-validation-sets-be/", "published_at": "2022-01-19T21:34:27+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a144a4e92", "title": "SageMaker Studio Lab: free AWS for learners", "url": "https://francescopochetti.com/sagemaker-studio-lab-free-aws-for-learners/", "published_at": "2021-12-07T19:39:49+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a14f9fc7a", "title": "IceVision + SAHI: democratise small object detection", "url": "https://francescopochetti.com/icevision-sahi-democratise-small-object-detection/", "published_at": "2021-12-01T10:07:57+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a15ab51e2", "title": "Fine-tune a DeepFake video classifier: PyTorchVideo, Lightning, W&B, and Amazon SageMaker in action", "url": "https://francescopochetti.com/fine-tune-a-deepfake-video-classifier-pytorchvideo-lightning-wb-and-amazon-sagemaker-in-action/", "published_at": "2021-11-14T21:03:36+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a163a7bc6", "title": "EasyOCR vs Tesseract vs Amazon Textract: an OCR engine comparison", "url": "https://francescopochetti.com/easyocr-vs-tesseract-vs-amazon-textract-an-ocr-engine-comparison/", "published_at": "2021-07-28T21:12:08+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a170a1e85", "title": "Developing inside a Docker Container in Visual Studio Code", "url": "https://francescopochetti.com/developing-inside-a-docker-container-in-visual-studio-code/", "published_at": "2021-07-15T09:55:43+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a17b07f0d", "title": "IceVision meets AWS: detect LaTeX symbols in handwritten math and deploy with Docker on Lambda", "url": "https://francescopochetti.com/icevision-meets-aws-detect-latex-symbols-in-handwritten-math-and-deploy-with-docker-on-lambda/", "published_at": "2021-01-21T08:10:49+00:00" }, { "id": "01a0c50d-a40f-7072-9d17-526a18625f37", "title": "Keypoint Detection with IceVision: my first contribution to open-source", "url": "https://francescopochetti.com/keypoint-detection-with-icevision-open-sourcing-computer-vision/", "published_at": "2020-11-30T13:01:29+00:00" }, { "id": "01a0c51c-939a-70e7-a874-ca626e1568d4", "title": "Deploying a Fashion-MNIST web app with Flask and Docker", "url": "https://francescopochetti.com/deploying-a-fashion-mnist-web-app-with-flask-and-docker/", "published_at": "2020-09-21T21:02:14+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cab43d391", "title": "DL for Computer Vision (Justin Johnson – University of Michigan): learning pills", "url": "https://francescopochetti.com/dl-for-computer-vision-justin-johnson-university-of-michigan-learning-pills/", "published_at": "2020-09-12T22:46:50+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cabe23dd9", "title": "A visual deep-dive into the building blocks of MobileNetV3", "url": "https://francescopochetti.com/a-visual-deep-dive-into-the-building-blocks-of-mobilenetv3/", "published_at": "2020-08-24T19:05:06+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cac3e4c75", "title": "Tabular Deep Leaning: TabNet deep-dive", "url": "https://francescopochetti.com/tabular-deep-leaning-tabnet-deep-dive/", "published_at": "2020-07-21T12:51:34+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cacb0eed9", "title": "PetFinder: a blend of Machine and Deep Learning approaches. CatBoost and MXNet in SageMaker Studio", "url": "https://francescopochetti.com/petfinder-a-blend-of-machine-and-deep-learning-approaches-catboost-and-mxnet-in-sagemaker-studio/", "published_at": "2020-06-12T11:12:15+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cad543e1f", "title": "Plant Pathologies: fastai’s wonders in Computer Vision", "url": "https://francescopochetti.com/plant-pathologies-fastais-wonders-in-computer-vision/", "published_at": "2020-04-22T12:31:03+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cae471164", "title": "Detecting machine VS human-generated Wikipedia articles with fastai", "url": "https://francescopochetti.com/detecting-machine-vs-human-generated-wikipedia-articles-with-fastai/", "published_at": "2020-04-04T10:37:35+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4caea97351", "title": "COVID-19: end-to-end analytics with AWS Glue, Athena and QuickSight", "url": "https://francescopochetti.com/covid-19-end-to-end-analytics-with-aws-glue-athena-and-quicksight/", "published_at": "2020-03-10T16:04:22+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4caf13a480", "title": "Whitening a black box: how to interpret a ML model", "url": "https://francescopochetti.com/whitening-a-black-box-how-to-interpret-a-ml-model/", "published_at": "2020-02-20T15:57:11+00:00" }, { "id": "01a0c51c-939b-73f4-b747-0a4cafb69c46", "title": "Fast Neural Style Transfer: deploying PyTorch models to AWS Lambda", "url": "https://francescopochetti.com/fast-neural-style-transfer-deploying-pytorch-models-to-aws-lambda/", "published_at": "2020-01-21T20:24:12+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786824a172a5", "title": "Recommend Expedia hotels with Amazon Personalize: the magic of Hierarchical RNNs", "url": "https://francescopochetti.com/recommend-expedia-hotels-with-amazon-personalize-the-magic-of-hierarchical-rnns/", "published_at": "2019-12-19T13:10:20+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-7868254d324e", "title": "Deploying a pretrained GPT-2 model on AWS", "url": "https://francescopochetti.com/deploying-a-pretrained-gpt-2-model-on-aws/", "published_at": "2019-12-04T19:43:49+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-7868258ec64e", "title": "Fast Neural Style Transfer: training the model", "url": "https://francescopochetti.com/fast-neural-style-transfer-training-the-model/", "published_at": "2019-09-29T18:40:39+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786825d99096", "title": "Fast Neural Style Transfer: deploying PyTorch models to Amazon SageMaker", "url": "https://francescopochetti.com/fast-neural-style-transfer-sagemaker-deployment/", "published_at": "2019-09-05T13:51:15+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-78682675a445", "title": "How to build an expense tracker with Amazon Textract", "url": "https://francescopochetti.com/how-to-build-an-expense-tracker-with-aws-textract/", "published_at": "2019-08-03T09:31:28+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786826a3040d", "title": "SageMaker Hyper-Parameter Optimization: classify heartbeat anomalies from stethoscope audio", "url": "https://francescopochetti.com/sagemaker-hyper-parameter-optimization-classify-heartbeat-anomalies-from-stethoscope-audio/", "published_at": "2019-06-19T16:08:55+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786826f0f25c", "title": "VisualNeurons.com, painting videos with S3, Cognito, Lambda, SES and AWS AI", "url": "https://francescopochetti.com/visualneurons-com-painting-videos-with-s3-cognito-lambda-ses-and-aws-ai/", "published_at": "2019-05-25T10:45:13+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786827887228", "title": "VisualNeurons.com, running Neural Style Transfer on AWS", "url": "https://francescopochetti.com/visualneurons-com-running-neural-style-transfer-on-aws/", "published_at": "2019-04-24T18:17:46+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786827c6456c", "title": "Extreme label imbalance: when you measure the minority class in basis points", "url": "https://francescopochetti.com/extreme-label-imbalance-when-you-measure-the-minority-class-in-basis-points/", "published_at": "2019-03-14T20:05:20+00:00" }, { "id": "01a0c527-bbf5-7301-8d65-786828c2772f", "title": "Is this movie a thriller? Exposing a SageMaker Deep Learning model in an end-to-end serverless Web Application", "url": "https://francescopochetti.com/is-this-movie-a-thrillerinvoking-a-sagemaker-deep-learning-model-in-an-end-to-end-serverless-web-application/", "published_at": "2019-03-05T12:25:25+00:00" } ] posts Claim your blog
Back to francescopochetti.com
Blog · corpus.blog/blogs/francescopochetti.com/posts

francescopochetti.com

francescopochetti.com

2025

Building Visual Neurons: image + video generation with Next.js, Replicate, and Amazon Bedrockoriginal ↗
11 Nov 2025

2024

Byte Pair Encoding: building the GPT tokenizer with Karpathyoriginal ↗
17 Mar 2024
CUDA for Python Programmersoriginal ↗
24 Feb 2024

2023

LogoNet: the journey to an AWS-powered cloud application running ControlNet on SageMaker async endpointsoriginal ↗
4 Jun 2023
A visual deep dive into the Transformer’s architecture: turning Karpathy’s masterclass into picturesoriginal ↗
9 Mar 2023
Fastai Course Part 2 2022: Understanding CallBacksoriginal ↗
17 Jan 2023

2022

Amazon SageMaker Shadow Deployment: Semantic Segmentation from the HuggingFace Huboriginal ↗
27 Dec 2022
Benchmarking TorchVision ResNet18 on EC2 NVIDIA GPU with TensorRT and Amazon SageMaker Neooriginal ↗
30 Nov 2022
Benchmarking TorchVision ResNet18 on Amazon SageMaker CPU, GPU, and Inferentia instances (with a Neo twist)original ↗
12 Nov 2022
Neural Magic: Training YoloV5 with Sparse Transfer Learning and deploying to Amazon SageMaker with a custom Docker containeroriginal ↗
31 Aug 2022
Training and Deploying a fully Dockerized License Plate Recognition app with IceVision, Amazon Textract and FastAPIoriginal ↗
26 Jun 2022
Blurry faces: Training, Optimizing and Deploying a segmentation model on Amazon SageMaker with NVIDIA TensorRT and NVIDIA Tritonoriginal ↗
22 May 2022
How much data should you allocate to training and validation?original ↗
19 Jan 2022

2021

SageMaker Studio Lab: free AWS for learnersoriginal ↗
7 Dec 2021
IceVision + SAHI: democratise small object detectionoriginal ↗
1 Dec 2021
Fine-tune a DeepFake video classifier: PyTorchVideo, Lightning, W&B, and Amazon SageMaker in actionoriginal ↗
14 Nov 2021
EasyOCR vs Tesseract vs Amazon Textract: an OCR engine comparisonoriginal ↗
28 Jul 2021
Developing inside a Docker Container in Visual Studio Codeoriginal ↗
15 Jul 2021
IceVision meets AWS: detect LaTeX symbols in handwritten math and deploy with Docker on Lambdaoriginal ↗
21 Jan 2021

2020

Keypoint Detection with IceVision: my first contribution to open-sourceoriginal ↗
30 Nov 2020
Deploying a Fashion-MNIST web app with Flask and Dockeroriginal ↗
21 Sept 2020
DL for Computer Vision (Justin Johnson – University of Michigan): learning pillsoriginal ↗
12 Sept 2020
A visual deep-dive into the building blocks of MobileNetV3original ↗
24 Aug 2020
Tabular Deep Leaning: TabNet deep-diveoriginal ↗
21 Jul 2020
PetFinder: a blend of Machine and Deep Learning approaches. CatBoost and MXNet in SageMaker Studiooriginal ↗
12 Jun 2020
Plant Pathologies: fastai’s wonders in Computer Visionoriginal ↗
22 Apr 2020
Detecting machine VS human-generated Wikipedia articles with fastaioriginal ↗
4 Apr 2020
COVID-19: end-to-end analytics with AWS Glue, Athena and QuickSightoriginal ↗
10 Mar 2020
Whitening a black box: how to interpret a ML modeloriginal ↗
20 Feb 2020
Fast Neural Style Transfer: deploying PyTorch models to AWS Lambdaoriginal ↗
21 Jan 2020

2019

Recommend Expedia hotels with Amazon Personalize: the magic of Hierarchical RNNsoriginal ↗
19 Dec 2019
Deploying a pretrained GPT-2 model on AWSoriginal ↗
4 Dec 2019
Fast Neural Style Transfer: training the modeloriginal ↗
29 Sept 2019
Fast Neural Style Transfer: deploying PyTorch models to Amazon SageMakeroriginal ↗
5 Sept 2019
How to build an expense tracker with Amazon Textractoriginal ↗
3 Aug 2019
SageMaker Hyper-Parameter Optimization: classify heartbeat anomalies from stethoscope audiooriginal ↗
19 Jun 2019
VisualNeurons.com, painting videos with S3, Cognito, Lambda, SES and AWS AIoriginal ↗
25 May 2019
VisualNeurons.com, running Neural Style Transfer on AWSoriginal ↗
24 Apr 2019
Extreme label imbalance: when you measure the minority class in basis pointsoriginal ↗
14 Mar 2019
Is this movie a thriller? Exposing a SageMaker Deep Learning model in an end-to-end serverless Web Applicationoriginal ↗
5 Mar 2019