Cortex LabsCortex Project Updates #121
Technical Updates 🤖Machine-Learning(ML), CortexVirtual Machine(CVM)Complete the function of coco, voc, cifar10 dataset loading;circom generator RightShift and MulScalar parameter alignment. Type annotations, Symbol type confusion fixes;Generate graph optimization, lose invalid nodes after topological sorting. scalar is integrated into op as an attribute, the original parameter is still output to input.json, fix circom compilation repair;Increase NLP model training, long sequence numerical pr...
Cortex LabsCortex Project Updates #126
ZKML2024-01-02 Publish [ZKML release 1.0.1] (https://github.com/CortexFoundation/tachikoma/commits/zkml-v1.0.1/) In this release, we implemented the general ZKML framework and passed the mnist model for test. 2024-01-16 Publish fixed commits for ZKML Fixed some classification models' quantization bugs, including mobilenet, squeezenet, etc.ZKCVMThe Cortex team has been closely following the latest zkEVM technology, which is one of the goals in the ZK trilogy (zkTX -> zkVM -> zkML). 2024-0...
Cortex LabsCortex Project Updates #111
Technical Updates 🤖Machine-Learning(ML), CortexVirtual Machine(CVM) & Model RepresentationTool(MRT) R&D MRT/CVMMRT updated the accuracy test code, adjusted and tested the model output results of the sampling dataAdjust the parameter setting of Calibrate Pass: absmax, fix the problem of FuseBatchnormRefactored the Quantiza Pass code and fixed the Infer Precision issueFixed the FixPoint Infer Type problem and added subgraph verification functionThe model parameters are adjusted to Int8, which ...
First decentralized world computer capable of running AI 🤖 and AI-powered Dapps ⚒. MainNet is out. Go #BUILD!🔥
Cortex LabsCortex Project Updates #121
Technical Updates 🤖Machine-Learning(ML), CortexVirtual Machine(CVM)Complete the function of coco, voc, cifar10 dataset loading;circom generator RightShift and MulScalar parameter alignment. Type annotations, Symbol type confusion fixes;Generate graph optimization, lose invalid nodes after topological sorting. scalar is integrated into op as an attribute, the original parameter is still output to input.json, fix circom compilation repair;Increase NLP model training, long sequence numerical pr...
Cortex LabsCortex Project Updates #126
ZKML2024-01-02 Publish [ZKML release 1.0.1] (https://github.com/CortexFoundation/tachikoma/commits/zkml-v1.0.1/) In this release, we implemented the general ZKML framework and passed the mnist model for test. 2024-01-16 Publish fixed commits for ZKML Fixed some classification models' quantization bugs, including mobilenet, squeezenet, etc.ZKCVMThe Cortex team has been closely following the latest zkEVM technology, which is one of the goals in the ZK trilogy (zkTX -> zkVM -> zkML). 2024-0...
Cortex LabsCortex Project Updates #111
Technical Updates 🤖Machine-Learning(ML), CortexVirtual Machine(CVM) & Model RepresentationTool(MRT) R&D MRT/CVMMRT updated the accuracy test code, adjusted and tested the model output results of the sampling dataAdjust the parameter setting of Calibrate Pass: absmax, fix the problem of FuseBatchnormRefactored the Quantiza Pass code and fixed the Infer Precision issueFixed the FixPoint Infer Type problem and added subgraph verification functionThe model parameters are adjusted to Int8, which ...
First decentralized world computer capable of running AI 🤖 and AI-powered Dapps ⚒. MainNet is out. Go #BUILD!🔥
Subscribe to Cortex Labs
Subscribe to Cortex Labs
Share Dialog
Share Dialog
<100 subscribers
<100 subscribers
Machine-Learning(ML), CortexVirtual Machine(CVM) & Model RepresentationTool(MRT) R&D
MRT/CVM
MRT, TVM and Circom unified representation structure, where MRT uses built-in InferType and InferShape
Add MRT operator to Circom's parser, and adapt to the shape adaptation of operators such as clip
Increase the export of Circom input parameters. The tested resnet data input size exceeds 100M. Consider integrating some parameters into the ZKP operator to reduce input consumption
Adjust the type check in MRT FixPoint, reduce TVM adaptation, and increase shaping check
Increase the type annotation of MRT Symbol, adjust the representation structure, and enhance the readability of the code
ZkRollup
The WitnessGenerator module adjusts some storage-related interfaces and adds merkle_state to generate proofs
Refactored the execution flow of StateKeeper: PendingBlock synchronously initializes StateKeeper and WitnessGenerator
Implement the L1Watcher module to monitor the data on the chain
Mempool handles unconfirmed events
ZKSync - era research (supporting L2 smart contracts), newly open-sourced part of the code of the GPU prover
Cortex Full Node
Machine-Learning(ML), CortexVirtual Machine(CVM) & Model RepresentationTool(MRT) R&D
MRT/CVM
MRT, TVM and Circom unified representation structure, where MRT uses built-in InferType and InferShape
Add MRT operator to Circom's parser, and adapt to the shape adaptation of operators such as clip
Increase the export of Circom input parameters. The tested resnet data input size exceeds 100M. Consider integrating some parameters into the ZKP operator to reduce input consumption
Adjust the type check in MRT FixPoint, reduce TVM adaptation, and increase shaping check
Increase the type annotation of MRT Symbol, adjust the representation structure, and enhance the readability of the code
ZkRollup
The WitnessGenerator module adjusts some storage-related interfaces and adds merkle_state to generate proofs
Refactored the execution flow of StateKeeper: PendingBlock synchronously initializes StateKeeper and WitnessGenerator
Implement the L1Watcher module to monitor the data on the chain
Mempool handles unconfirmed events
ZKSync - era research (supporting L2 smart contracts), newly open-sourced part of the code of the GPU prover
Cortex Full Node
pepple MemTableStopWritesThreshold fix
fix off-by-one in ipc endpoint length check
some funcs added and fix about pee
BSC Station Partnership
Cortex NFT Gallery internal test
Latest Cortex explorer test
Cortex NFT Gallery Guide
Research on P2E
Continue to monitor popular market ventures and examine project logic and Web3, NFT & Defi Tokenomic models
About Cortex 😇
Cortex’s main mission is to provide state-of-the-art machine-learning models on the blockchain in which users can infer using smart contracts on the Cortex blockchain. One of Cortex’s goals also includes implementing a machine-learning platform that allows users to post tasks on the platform, and submit AI DApps (Artificial Intelligence Decentralized Applications).
Cortex is the only public blockchain that allows the execution of nontrivial AI algorithms on the blockchain. MainNet has launched. Go build!
TestNet
Social Media
Telegram Groups
pepple MemTableStopWritesThreshold fix
fix off-by-one in ipc endpoint length check
some funcs added and fix about pee
BSC Station Partnership
Cortex NFT Gallery internal test
Latest Cortex explorer test
Cortex NFT Gallery Guide
Research on P2E
Continue to monitor popular market ventures and examine project logic and Web3, NFT & Defi Tokenomic models
About Cortex 😇
Cortex’s main mission is to provide state-of-the-art machine-learning models on the blockchain in which users can infer using smart contracts on the Cortex blockchain. One of Cortex’s goals also includes implementing a machine-learning platform that allows users to post tasks on the platform, and submit AI DApps (Artificial Intelligence Decentralized Applications).
Cortex is the only public blockchain that allows the execution of nontrivial AI algorithms on the blockchain. MainNet has launched. Go build!
TestNet
Social Media
Telegram Groups
No activity yet