Release FL on SGX V1.0 #882
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
duanbing
commented
Aug 24, 2021
•
duanbing
commented
- Mutual RA-TLS by GRPC between worker and ps, worker and worker;
- Upgrade graphene to graming for production-ready;
- New sample codes
Co-authored-by: root <[email protected]>
* feat(raw_data): add parameter validation for data portal job * fix pylint
* fix follower restart index missing * fix follower restart index missing
* fix(psi): PSI failed when CPU_LIMIT < 4 * remove unused package
* feat(tree): select label * feat(tree): fix * feat(tree): add label_feild in script Co-authored-by: 杭卫强 <[email protected]>
* Fix default value double encode in csv * Fix default value double encode in csv * Fix default value double encode in csv * debug print missing value
* feat(tree): add eval metrics * feat(tree): add load_model_name Co-authored-by: 杭卫强 <[email protected]>
* fix(tree): remove x_names=none * fix(tree): fix Co-authored-by: 杭卫强 <[email protected]>
Co-authored-by: 杭卫强 <[email protected]>
* fix(common): do not modify value when convert tf example to dict * fix(tfrecord): fix convert tfrecord to dict * fix: add comment
Co-authored-by: sft <[email protected]>
* feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): append more metrics to nn model * feat(metrics): fix comments * feat(metrics): fix comments
Co-authored-by: gejielun <[email protected]>
Co-authored-by: gezhengqiang <[email protected]>
* add meituan hdfs IO feature * remove loader.env.CLASSPATH, it's loaded from host now. * Pull code script adapt Meituan viewfs hdfs --------- Co-authored-by: huangxu17 <[email protected]> Co-authored-by: gejielun <[email protected]>
…nents (#1092) * feat(sgx): support tensorflow remote attestation in list * fix(sgx): improve EPC cost * fix(sgx): fix memory leak of json_engine * fix(sgx):misspell Ps->ps
* Update estimator.py: add marvell related * Update trainer_worker.py: add arguments * Add privacy via upload: add protects and attacks * Update __init__.py: add privacy into package * Update marvell.py：Add comments * Update marvell.py: add comments on zero division problems * Update norm_attack.py: add comments * Update emb_attack.py: add comments * Update discorloss.py: add paper * Update estimator.py: add comments * Update trainer_worker.py:add comments * Update discorloss.py:add comments * adjust structure move the privacy protection related codes to /privacy/splitnn * Delete fedlearner/privacy/discorloss.py * Delete fedlearner/privacy/emb_attack.py * Delete fedlearner/privacy/marvell.py * Delete fedlearner/privacy/norm_attack.py * Update estimator.py: label protection
* feat(sgx): support tensorflow remote attestation in list * fix(sgx): improve EPC cost * fix(sgx): fix memory leak of json_engine * fix(sgx):misspell Ps->ps * fix(sgx): fix get_token * feat(sgx): put all python packages into trusted_files
* Meituan HDFS Access Without Proxy for Kerberos Authentication * Meituan HDFS Access Without Proxy for Kerberos Authentication --------- Co-authored-by: huangxu17 <[email protected]>
* Meituan HDFS Access Without Proxy for Kerberos Authentication * Meituan HDFS Access Without Proxy for Kerberos Authentication * 1. FedLearner Framework and Core Dependency RA-TLS Configuration To ensure code consistency and implement hash signature-based remote attestation for the FedLearner framework and its core dependencies, an environment variable has been introduced in Gramine that prevents the generation of .pyc files. 2. Gramine Template Configuration for FedLearner The Gramine template configuration has been updated to include the code locations for the FedLearner framework and some essential dependencies. This addition facilitates bidirectional remote attestation between parties. 3. Meituan HDFS File Path Management Optimization The code responsible for handling file paths in Meituan's Hadoop Distributed File System (HDFS) has been migrated from the main entry point to be processed by the master node. This change aims to prevent file read and write conflicts that could arise from multiple workers operating simultaneously. * Fix gramine allowed files lib risk, add sub path to allowed files * Regarding the TensorFlow API, it allows the addition of specific files that change with each compilation to the Gramine configuration, rather than the entire API folder. --------- Co-authored-by: huangxu17 <[email protected]>
Co-authored-by: huangxu17 <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.