Tensorflowemplois

...basse 40×8…”) Génération automatique d’un modèle 3D paramétrique Production de plans 2D cotés, schémas d’assemblage et fiches de débit Export des fichiers vers DWG / PDF / STEP / IFC Collaboration avec le commanditaire pour valider et améliorer le prototype Profil recherché : Étudiant ou jeune développeur avec passion pour l’IA, la 3D et la CAO Connaissances souhaitées : Python / IA (PyTorch, TensorFlow) et/ou CAO 3D (Blender, FreeCAD, OpenCascade, Tekla, AutoCAD) Autonome, motivé et curieux Matériel requis : juste un PC et connexion Internet Conditions : Pas de salaire immédiat : rémunération via 10 % des revenus des abonneme...

- Créer une application web permettant de visualiser des données de ventes. - Intégrer un module IA pour prédire les tendances de consommation. - Technologies préférées : Python, Django, Power BI, TensorFlow. - Livraison attendue : prototype fonctionnel en 2 mois. - Compétences requises : Data Science, BI, développement web.

... reconstruire la géométrie, nettoyer le maillage et, si possible, produire un format standard (OBJ/FBX/GLB). • le code commenté, les modèles entraînés et un court guide d’installation. • une démonstration sur plusieurs scans fournis pour valider la qualité de sortie (maillage étanche, nombre de polygones raisonnable, artefacts minimaux). Je reste ouvert à vos choix techniques – PyTorch, TensorFlow, Blender Python, MeshLab, Cloud services… tant que le pipeline est reproductible. Si vous avez déjà travaillé avec des réseaux de type Neural Implicit Surface, point-cloud to mesh ou des approches similaires, précisez-le ; c’est un vrai plus. F...

...l’optimisation d’un modèle de classification sous Python, tout en menant l’analyse exploratoire nécessaire pour vous livrer des insights clairs. Vos jeux de données numériques, textuels et même vos images seront traités dans un pipeline cohérent : nettoyage, préparation, sélection de variables, entraînement, évaluation puis export du modèle prêt à être déployé. J’utilise couramment pandas, scikit-learn, TensorFlow/Keras et Jupyter ; je peux également intégrer votre stack existante si besoin. Livrables sous 24 h : • Notebook ou script Python entièrement commenté • Rapport concis sur les performances (m...

Je souhaite lancer un premier MVP d’application web capable d’exploiter la vidéo pour aider les athlètes à corriger leur technique sur trois mouvements clefs – squat, développé couché et deadlift. Le flux idéal est simple : l...fonctionnel hébergé en ligne (front + back) • Démonstration de la détection de posture, calcul du score et génération du feedback personnalisé • Module basique de suivi des progrès (graphiques ou tableau synthétique) • Interface bilingue FR / EN • Documentation d’installation et de prise en main Merci d’indiquer les outils ou frameworks IA/vision par ordinateur que vous maîtrisez (Tens...

...activation sigmoid, orienté « équilibre entre précision et complexité ». • Création d’un générateur de données d’entraînement : swing points + filler points pour reproduire les fluctuations de marché et augmenter le jeu de données. • Intégration des données OHLCV comme indicateur de base, avec la possibilité d’ajouter vos propres features si cela améliore la détection. • Scripts clairs (TensorFlow, Keras ou PyTorch) et commentaires détaillés pour que je puisse ré-entraîner ou ajuster le modèle. • Rapport succinct expliquant la méthodologie : préparation des donn&eacu...

...activation sigmoid, orienté « équilibre entre précision et complexité ». • Création d’un générateur de données d’entraînement : swing points + filler points pour reproduire les fluctuations de marché et augmenter le jeu de données. • Intégration des données OHLCV comme indicateur de base, avec la possibilité d’ajouter vos propres features si cela améliore la détection. • Scripts clairs (TensorFlow, Keras ou PyTorch) et commentaires détaillés pour que je puisse ré-entraîner ou ajuster le modèle. • Rapport succinct expliquant la méthodologie : préparation des donn&eacu...

...activation sigmoid, orienté « équilibre entre précision et complexité ». • Création d’un générateur de données d’entraînement : swing points + filler points pour reproduire les fluctuations de marché et augmenter le jeu de données. • Intégration des données OHLCV comme indicateur de base, avec la possibilité d’ajouter vos propres features si cela améliore la détection. • Scripts clairs (TensorFlow, Keras ou PyTorch) et commentaires détaillés pour que je puisse ré-entraîner ou ajuster le modèle. • Rapport succinct expliquant la méthodologie : préparation des donn&eacu...

...un site intégrant des fonctionnalités d’intelligence artificielle. L’objectif est d’obtenir une boutique complète, fluide et sécurisée, capable de dialoguer avec mes visiteurs via un chatbot intelligent et de générer automatiquement certains contenus (fiches produit, articles, FAQ, etc.). Technologies privilégiées : HTML, CSS, JavaScript, React ou équivalent pour le front-end ; OpenAI API ou TensorFlow pour les briques IA. Je reste ouvert à vos suggestions sur les outils et frameworks tant qu’ils permettent un résultat fiable, performant et évolutif. Livrables attendus : • Maquettes et parcours utilisateur (Figma, Adobe XD ou similaire) • Dévelop...

...projet (Agile, Scrum, Kanban). - Capacité à élaborer des plans de projet détaillés, définir les objectifs, les délais et les ressources nécessaires. 2. **Compétences Techniques** : - Compréhension des principes de base du développement logiciel, y compris les technologies web et les API RESTful. - Familiarité avec les outils de traitement d'images et de machine learning (OpenCV, TensorFlow) est un plus. - Connaissance des systèmes de diffusion vidéo et des protocoles de communication (RTMP, HTTP). 3. **Communication et Collaboration** : - Excellentes compétences en communication écrite et verbale pour interagir avec des équipes tech...

1 Recherche d'outils: - Trouver des bibliothèques comme OpenCv pour le traitement d'images et TensorFlow pour le machine learning. 2 Développement du modèle: - Collecter des exemples de publicités pour entrainer le modèles - Utiliser ces exemples pour enseigner à l'IA comment reconnaitre les publicités dans un flux vidéo 3 Intégration: - Créer une API qui permettra à d'autres parties du système de demander des analyse de flux vidéo Livrables: Modèle de reconnaissance de publicités API d'analyse de flux vidéo

Bonjour, Je suis à la recherche d’un développeur web freelance pour créer une application web et mobile intégrant un système d’abonnement et de paiement en ligne. Présentation du projet : L’application sera un assistant nutritionnel intelligent, capable d’analyser l...principales : Reconnaissance automatique des aliments via IA Calcul des calories et des valeurs nutritionnelles Suivi des repas et des objectifs Système d’abonnement avec paiements récurrents (Stripe, PayPal, Google/Apple Pay) Interface intuitive et design attractif Compétences requises : Front-end : React, Vue.js ou autre Back-end : Node.js, Django, Laravel, etc. IA & traitement d’image (TensorFlow, Open...

...AI response time < 2 seconds for classification. Scalability Handle 10,000+ patient records efficiently. Availability 99.9% uptime, cloud-based deployment. Compliance GDPR, HIPAA, and other medical data regulations. 6. Technology Stack Component Technology Frontend React, Angular, or Vue.js Backend Node.js, Python (FastAPI/Django) Database PostgreSQL, MongoDB (for patient records) AI Model TensorFlow, Scikit-learn, PyTorch Security OAuth 2.0, JWT Authentication EHR Integration FHIR API, HL7 7. Project Timeline Phase Tasks Duration Phase 1: Planning Requirement analysis, system design 2 weeks Phase 2: Development Backend, AI model training, frontend UI 8 weeks Phase 3: Integration EHR system integration, API development 4 weeks Phase 4: Testing Unit testing, AI model valida...

...modèles d’intelligence artificielle open source pour la reconnaissance alimentaire. Le projet repose sur des frameworks comme TensorFlow, YOLO ou Fastai et des ensembles de données tels que Food-101. L'objectif est de développer une application performante, user-friendly, capable d'identifier des aliments à partir d'images et de fournir des informations nutritionnelles ou des recommandations. Responsabilités Adapter et entraîner des modèles open source pour la reconnaissance alimentaire. Développer une application mobile native ou cross-platform (iOS et Android) en utilisant Swift/Kotlin ou Flutter/React Native. Intégrer le modèle IA à l'application mobile via TensorFlow...

...Données : Intégrer un appareil de capture vidéo à un logiciel de traitement de données pour l’analyse. Technologies et Outils TensorFlow : Pour l’analyse vidéo et la reconnaissance d’images. OpenCV : Pour le traitement d’images et la vision par ordinateur. PyTorch : Pour le développement de modèles de deep learning flexibles. Keras : Pour créer et entraîner des modèles de réseaux de neurones. Scikit-learn : Pour l’analyse de données et les prédictions statistiques. Compétences Requises Maîtrise de TensorFlow ou équivalent (par exemple, PyTorch) Maîtrise d’OpenCV ou équivalent (par exemple, SimpleCV) ...

...Prévisualisation des projets : • Génération automatique de liens de prévisualisation pour permettre aux clients de consulter les sites/applications en cours de développement. Exigences techniques : • Développement en JavaScript, HTML/CSS, avec un framework moderne (ex: React, Angular). • L’intelligence artificielle peut être développée à l’aide de Python ou un autre langage adapté à l’IA (TensorFlow, OpenAI, etc.). • Le logiciel doit être responsive, c’est-à-dire qu’il doit s’adapter à tout type d’écran. • Base de données pour stocker les projets, préférences utilisateurs, etc. ...

...data to train the algorithm. - **Validation**: Test the algorithm with validation data to evaluate its accuracy. - **Deployment**: Integrate the algorithm into the application for real-time predictions. ## Technologies and Tools for Analytics 1. **Backend** - **Language**: Python is ideal for data processing and machine learning. - **Machine Learning Frameworks**: Scikit-learn, TensorFlow, PyTorch. - **Database**: PostgreSQL or MySQL for storing school data. 2. **Frontend** - **Frameworks**: React.js, Vue.js, Angular for creating an interactive user interface. - **Visualization Libraries**: D3.js, , Plotly for displaying interactive graphs and tables. ## Development Plan for Analytics ### 1. Research and Analysis Phase - Identify available data and sp...

Nous recherchons une personne pour nous aider à créer le prototype d'une application SaaS basée sur Tensorflow. La mission consisterait à : - définir les pré-requis techniques - installer l'environnement et documenter la procédure de mise en œuvre - définir / documenter des procédures d'apprentissage de classification de texte

Nous avons fait appel a un particulier afin de developper un programme de reconnaissance d'objet avec Tensorflow, il y a aussi de la commande de moteurs pas à pas. Il a réussi a faire un programme, cependant nous avons besoin de le dubugger, d'augmenter sa fiabilité et sa consommation d'énergie

Développer une application web pour la prédication de l’évolution des malades du Covid-19 dans un pays avec Tensorflow, et React.js et Mysql.

...production-grade product, not a toy demo. The long-term engagement is several weeks to months, with competitive compensation. Skills We're Looking For Strong experience with deep learning for face/video generation (GANs, diffusion, NeRF, or similar) Hands-on experience with models like SadTalker, Wav2Lip, MuseTalk, LivePortrait, Thin-Plate Spline, FOMM, or similar Proficiency in Python, PyTorch/TensorFlow Experience with real-time streaming (WebSocket, WebRTC, RTMP, GStreamer) Ability to optimize inference for real-time performance (TensorRT, ONNX, model quantization) Bonus: experience with TTS pipelines (Coqui, Bark, XTTS, ElevenLabs integration) How to Enter Build the POC following the specs above Record your live demo (screen + mic, showing real-time sync) Upload you...

...and interpretability both matter. Here’s what I need from you: • A brief data-exploration notebook that highlights key correlations, missing-value handling, and basic visuals. • Feature engineering tailored to the data’s domain (scaling, encoding, derived metrics, etc.). • At least two supervised algorithms (for example, Gradient Boosting and Random Forest in scikit-learn, or an XGBoost/TensorFlow alternative) trained, cross-validated, and benchmarked. • A concise performance comparison using appropriate regression/classification metrics—whichever fits once you see the target variable. • The final, best-performing model saved in a reusable format (pickle/joblib or SavedModel). • A short read-me that explains: setup step...

...and interpretability both matter. Here’s what I need from you: • A brief data-exploration notebook that highlights key correlations, missing-value handling, and basic visuals. • Feature engineering tailored to the data’s domain (scaling, encoding, derived metrics, etc.). • At least two supervised algorithms (for example, Gradient Boosting and Random Forest in scikit-learn, or an XGBoost/TensorFlow alternative) trained, cross-validated, and benchmarked. • A concise performance comparison using appropriate regression/classification metrics—whichever fits once you see the target variable. • The final, best-performing model saved in a reusable format (pickle/joblib or SavedModel). • A short read-me that explains: setup step...

...(vector analysis). 3. State Transition: The algorithm must seamlessly transition from "Bulge Tracking" (closed eyes) to standard "Pupil/Iris Tracking" (open eyes) within the same 30-second video clip. 4. Mobile Optimization: The final module must be lightweight and optimized for real-time or near-real-time performance on iOS and Android devices. Candidate Profile • Expertise in OpenCV, MediaPipe, TensorFlow Lite, or CoreML. • Proven track record in Image Processing or Eye-Tracking projects. • Strong background in Motion Analysis and extracting signals from noisy video data. • Experience in delivering code ready for mobile integration. Application Instructions Please include the following in your proposal: • Examples of previous work in...

...project centres on building a production-ready medical image -classification pipeline that leverages modern deep-learning techniques. I have a labelled dataset and need end-to-end code that ingests the text, handles cleaning and tokenisation, and trains an accurate classifier. Python is the preferred language; The preprocessing must involve Quantum computing techniques using Pennylane. PyTorch, TensorFlow or another mainstream framework is fine as long as the solution is reproducible and easy to extend. Key deliverables: • Well-commented source code (data loading, model, training loop, evaluation) • Clear instructions to run training on a fresh machine (README or notebook) • Metrics report showing accuracy, precision, recall and F1 on a held-out set • ...

...(Finacle, Oracle, Temenos, or similar) │ ├── Developed reconciliation engines for payments │ ├── Implemented fraud detection systems │ └── Deployed systems in regulated environments (finance, healthcare, government) NICE TO HAVE: ├── Experience with Africa's Talking SMS gateway ├── Knowledge of Levenshtein distance or fuzzy matching algorithms ├── Experience with ML model deployment (TensorFlow, PyTorch) ├── Previous work with microservices architecture ├── Experience with Elasticsearch, Kibana (ELK stack) ├── Familiarity with banking regulations (data protection, PCI-DSS) └── Existing relationships with telco or bank technical teams WHAT WE ALREADY HAVE (The Assets You'll Integrate) We are NOT starting from scratch. The following assets are already bui...

suspicious activity detection using deep learning # Features Real-Time Detection: Monitors live camera feeds and detects suspicious activities instantly. Multi-Class Detection: Classifies activities such as hara...fighting, vandalism, and abuse. Alert System: Sends real-time notifications to authorities upon detecting suspicious activities. Dataset-Based Training: Trained on UCF-Crime, AVENUE Video Dataset, and Violent-Flows for high accuracy. Technologies Used YOLOv8: For real-time object and people detection. CNN-LSTM: For recognizing and classifying human activities. Deep Learning Frameworks: TensorFlow, PyTorch. Backend: Flask/Django (replace with your backend framework). Frontend: Minimalistic web interface for real-time monitoring. Alert System: Email (integrated w...

...lighting at the doorway • Automatic in/out logic that prevents double entries when someone lingers near the camera • Admin dashboard with search, manual override and CSV/XLS export • Installation guide plus brief training so I can manage enrolments myself • Source code or licence details, and clear instructions for future camera or user expansion Feel free to use OpenCV, DeepFace, TensorFlow, AWS Rekognition, or another proven toolset—just explain the choice and any recurring costs. I will supply the RTSP stream from the entrance camera and can provision a small Windows or Linux box on-site if you recommend an on-prem edge device. Once everything is running smoothly and the test group’s times are logging correctly for a full week, I&r...

...must be available both as a responsive web application and as native or cross-platform mobile apps, all drawing from a single codebase whenever practical to streamline future maintenance. Core expectations • Modern, engaging UI/UX that feels consistent across web, iOS and Android. • A secure, multi-tenant architecture so each school’s data remains isolated. • An AI layer (think Python, TensorFlow/-Lite, or similar) that plugs into the learning workflow—for example adaptive content or real-time feedback—without locking us into a specific vendor. • Real-time sync between devices, even on spotty school Wi-Fi, and offline caching for mobile. • Admin dashboard with role-based access, permissions and exportable analytics. • C...

...fast delivery (within 24 hours) - Long-term work possible if successful Scope of Work: - Diagnose deep learning pipeline issues - Fix model execution errors - Debug training / inference workflow - Resolve dependency or environment conflicts - Optimize pipeline stability - Ensure end-to-end execution works correctly - Provide brief documentation of fixes Technical Stack: - Python - PyTorch / TensorFlow - HuggingFace / Transformers - CUDA / GPU acceleration - Docker / Linux environment - API integration & Data preprocessing pipeline Requirements: - Strong experience in Deep Learning production workflows - Experience debugging complex AI pipelines - Comfortable working under urgent timelines and ability to start immediately Timeline: Start: Immediately. Expected turnaround: ...

Veuillez vous inscrire ou vous connecter pour voir les détails.

Veuillez vous inscrire ou vous connecter pour voir les détails.

Veuillez vous inscrire ou vous connecter pour voir les détails.

Looking for a highly skilled retail, merchandising, and pricing Data Scientist with deep expertise in AI, Generative AI, and NLP, who has hands-on experience building the following and can walk through in-depth examples and solutions they have implemented across diverse real-world scenarios to drive scalable retail, merchandi...accuracy. Deployment & Integration: Integrate solutions with applications and data systems via APIs and web services, ensuring scalability and reliability. Research & Development of Emerging Technologies: Stay updated on the latest AI/ML advancements and explore opportunities to incorporate innovations into merchandising and pricing transformation initiatives. Frameworks & Tools: TensorFlow, PyTorch, OpenAI, LangChain, and other mod...

...methodology into clean, reproducible code. The core help I’m after is coding itself—covering the full pipeline from data preprocessing through model training to final evaluation and visualisation. I need datasets, well-documented Python scripts or notebooks that I can run end-to-end on my own machine (or a Colab instance). Expect to work with common libraries such as pandas, NumPy, PyTorch or TensorFlow, Hugging Face Transformers, plus Matplotlib or Seaborn for charts—use whichever combination best suits the objectives while keeping dependencies manageable. Deliverables
 • Data preprocessing module that loads the provided datasets, cleans them, applies any necessary tokenisation and splits them into train/validation/test sets. • Training script tha...

Looking for a highly skilled Data Scientist with deep expertise in AI and Generative AI to lead cutting-edge retail merchandising and pricing analytics initiatives, driving scalable forecasting, optimization, and intelligent ...-Implement optimization models to improve operational efficiency and maximize customer value. Manage the full machine learning lifecycle, including model monitoring, retraining, experiment tracking, and performance evaluation. -Develop and maintain CI/CD pipelines for reliable and scalable deployment of data science solutions. -Work with modern machine learning frameworks and tools such as TensorFlow, PyTorch, OpenAI, and LangChain. -Collaborate with cross-functional teams and integrate solutions with existing applications and data systems via APIs and web...

...common pesticide residues on fresh fruit and vegetable samples. My current lab setup streams CSV files over USB; if you have dealt with other device protocols, feel free to propose an efficient data-capture approach. Core requirements • A clean Python pipeline that parses the spectra, performs any necessary preprocessing (baseline correction, smoothing, normalization), and feeds the data into a TensorFlow model. • A well-documented training notebook + scripts so the model can be re-trained when new pesticides or produce types are added. • (Optional but welcome) a complementary computer-vision module. If you have experience with object detection, segmentation, or classic feature extraction, show me how you would fuse image cues with the spectral output to boo...

...can 1) recognise each image, 2) detect all relevant objects inside it, and 3) optionally segment those objects when that adds value to the overall result. The primary goal is accurate detection and classification, but I’d like the system to be flexible enough to switch on pixel-level segmentation whenever it boosts performance. Here is what I’m after: • A clean Python pipeline (PyTorch or TensorFlow preferred, with OpenCV for preprocessing) that ingests my custom images, trains suitable models, and exposes an easy inference script. • Clear evaluation: confusion matrices, mAP/IoU scores, and plots of loss/accuracy over epochs so I can judge progress at a glance. • A concise, publication-ready report written in LaTeX. I will supply the template; yo...

I need an AI-driven workflow that ingests raw video (MP4/MOV) and automatically follows a chosen object through every frame, delivering a clean alpha-masked output I can drop straight into my edit. The goal is to save me from manual rotoscoping; accuracy and speed are both critical. Here’s how I picture the collaboration: you build or adapt a model—using tools such as PyTorch, TensorFlow, Detectron2, or a proven OpenCV pipeline—that detects the target, keeps the mask tight even during fast motion, and exports either a keyed video or a PNG sequence. A small sample clip will be provided for you to demonstrate proof of concept; once results look solid we’ll run the system on the full batch. Deliverables • Python or command-line script (with all depe...

...AI-powered image-recognition tool that focuses exclusively on identifying everyday household items. The core requirement is clear: when the model sees a photo, it should reliably detect and classify objects such as chairs, tables, kettles, lamps, and similar items that people typically keep at home. Here is how I picture the workflow and final hand-off: • Model: A well-trained neural network (TensorFlow, PyTorch, or a comparable framework) tuned for object detection/classification. • Dataset handling: Either you assemble and label a suitable open-source dataset or guide me on licensing a ready-made one; in either case, the final dataset or clear reproducibility steps must be included. • Inference pipeline: A simple script or lightweight API endpoint so I can ...

...feeds or historical fault libraries later, so a clean, extensible data pipeline matters. Key deliverables • Cross-platform mobile app (iOS + Android) built with a modern stack—React Native, Flutter, or another framework you are comfortable with. • Fault-tree engine that mirrors Airbus AMM logic and lets me update procedures without redeploying the whole app. • Predictive module (Python/TensorFlow, PyTorch, or similar) that ranks probable troubleshooting branches based on past fixes. • Secure local/remote storage of maintenance logs, plus export in CSV or JSON for MIS upload. • Clear documentation and a short video demo showing the workflow on an A320 use-case. Acceptance criteria 1. Given a sample logbook entry “F/CTL PRIM1 FAULT...

...returns its class. • Documentation: concise README explaining setup, dependencies, and how to retrain with fresh data. Acceptance criteria 1. Minimum F1-score of 0.85 on the hold-out test set I will supply. 2. Reproducible environment ( or ). 3. Code delivered via private Git repository or ZIP file. Tools you might consider include Python 3.11, PyTorch or TensorFlow, HuggingFace Transformers, spaCy, and scikit-learn; feel free to propose alternatives if they achieve equal or better results. I will review interim results as soon as you have an initial baseline, then we can iterate on hyper-parameters, class imbalance handling, and deployment details....

...shots, slight rolls, and images with minimal vertical cues. 2. Batch processing and an interactive before-after comparison are mandatory. 3. Front-end must run in any modern browser; a minimal back-end is fine as long as uploads are secure and temporary. 4. Code must be clean and well-documented so I can extend it later. I have no fixed stack preference, so feel free to propose OpenCV, TensorFlow, or any combination of JavaScript (e.g. WebAssembly-compiled OpenCV), Python (FastAPI, Flask), or other technologies you’re comfortable with. Please outline: • The libraries and frameworks you’ll rely on. • Your approach to vanishing-point detection and homography estimation (e.g., RANSAC line clustering, deep-learning refinement, etc.). • Expected ...

...me: • Interactive graphics and videos must steal the show—they should illustrate key concepts, animate data, and invite the viewer to click, pause, or explore. • Avatars must look realistic, lip-sync flawlessly, and be easy to re-skin for future episodes. • The entire pipeline—from text input to final MP4—should run with minimal manual tweaking, whether you build it in Python with PyTorch/TensorFlow or orchestrate commercial generative-media APIs. Deliverables 1. A working proof-of-concept that accepts a text script, generates the realistic avatar narration, and stitches in AI-created graphics/video to produce a cohesive training module. 2. Source code plus clear setup and usage instructions. 3. A short sample episode built from one of my...

...machine-learning model (anomaly detection or sequence-based classification) optimised for on-device or near-edge execution. 2. Implement a decision layer that selects one of three responses—encrypt, quarantine, or alert—based on the model’s confidence score. 3. Provide clean, well-commented Python code (TensorFlow, PyTorch or scikit-learn are all acceptable) plus a short README explaining data preprocessing, hyper-parameters and how to port the model to an embedded runtime (e.g., TensorFlow Lite, ONNX). 4. Supply a small synthetic data set and demonstrate at least 90 % accuracy in distinguishing normal from suspicious activity during a live demo or recorded notebook. Acceptance criteria • Model trains and runs locally on a laptop within 10 min...

I need a lightweight image-classification prototype that cleanly separates iPads from iPhones. You will work in Google’s Teachable Machine so the end result can be demonstrated live to non-technical stakeholders and, if needed, exported for further use in TensorFlow or a Python pipeline later on. Data I will supply a mixed set of photos—my own device shots plus carefully curated stock images—so the training set covers varied angles, lighting, and backgrounds. Target performance The prototype should reach better than 90 % accuracy on fresh, unseen images. Please incorporate any practical tricks (augmentation, class-balance tweaks, transfer learning, etc.) that help hit this benchmark without overcomplicating the workflow. Workflow & knowledge transfer A...

...into Reinforcement Learning or NLP later, that flexibility will be a plus for the longer roadmap, but the immediate priority is Deep Learning mastery. The sessions will be delivered live (online or hybrid can be arranged), and I’ll rely on you to: • Shape a clear, week-by-week syllabus covering CNNs, RNNs, transformers, optimisation tricks, model interpretability and MLOps basics using Python, TensorFlow or PyTorch • Provide concise slide decks, hands-on notebooks (Jupyter/Colab) and at least three graded mini-projects that mirror industry use-cases • Guide learners through code reviews and Q&A, then wrap up with a capstone evaluation and feedback report All teaching material must be original or properly licensed, and ready for hand-off at the end ...

...training, validation, and test sets, with any necessary feature engineering you judge appropriate. I have no fixed preference on the final algorithm—linear models, tree ensembles, or a small neural network are all acceptable as long as they deliver solid predictive accuracy and are easy to retrain when I add more data. Please build the solution in standard Python tooling (pandas, scikit-learn, TensorFlow or PyTorch only if the accuracy gains justify it) and present the work in a Jupyter Notebook. Your notebook should walk me through: • data import, preprocessing, and exploratory visuals • model selection and cross-validated performance metrics • prediction of W/L ratio on unseen inputs • a short optimisation routine that searches the design space...

...dosage, batch number, expiry date, manufacturer, and any other legible data. • Return the result programmatically as a JSON object so it can be stored or sent to our API. Accuracy is more important than speed, but I still expect real-time feedback on focus and framing. You’re free to leverage mobile-friendly vision libraries (Google ML Kit, Tesseract, Vision Framework, etc.) or a custom TensorFlow-Lite model if that yields better results. Everything must run on-device; no cloud calls. Deliverables: 1. Full source code for the iOS and Android implementation (native, Flutter, or React Native—use what lets you hit the quality bar fastest). 2. A short read-me that explains build steps, dependencies, and the JSON schema. 3. Sample JSON output from at least t...

...machine learning model that reliably classifies each photo into the correct category. Your job is to design, train, and evaluate the full image-classification pipeline. You may build from scratch or fine-tune a proven architecture such as ResNet, EfficientNet, MobileNet, or a vision transformer—as long as the final model meets the accuracy targets we set together. Feel free to work in PyTorch or TensorFlow/Keras; I’m comfortable deploying either. What I’ll provide • A structured folder of training, validation, and test images • Category labels and a brief data dictionary • Access to a GPU instance if you need it What I need back 1. Clean, well-commented code (Jupyter notebook or Python scripts) that handles preprocessing, augmentation, trai...