From 625d1d258af85d1106c8472a5e4f0f43e65b78c3 Mon Sep 17 00:00:00 2001 From: Yasar Abbas Date: Mon, 18 Dec 2023 16:05:28 -0600 Subject: [PATCH] Add FedVSSL baseline (#2412) Co-authored-by: yan-gao-GY Co-authored-by: jafermarq Co-authored-by: Daniel J. Beutel --- baselines/fedvssl/.gitignore | 2 + baselines/fedvssl/LICENSE | 202 ++++++++++++++++ baselines/fedvssl/README.md | 227 ++++++++++++++++++ baselines/fedvssl/_static/FVSSL.png | Bin 0 -> 1012736 bytes baselines/fedvssl/fedvssl/__init__.py | 1 + baselines/fedvssl/fedvssl/client.py | 141 +++++++++++ baselines/fedvssl/fedvssl/conf/base.yaml | 32 +++ .../conf/mmcv_conf/finetuning/model_r3d18.py | 32 +++ .../r3d_18_ucf101/finetune_ucf101.py | 11 + .../finetuning/r3d_18_ucf101/test_ucf101.py | 11 + .../mmcv_conf/finetuning/runtime_ucf101.py | 137 +++++++++++ .../pretraining_runtime_kinetics.py | 34 +++ .../pretraining/pretraining_runtime_ucf.py | 134 +++++++++++ .../pretraining_for_kinetics.py | 36 +++ .../r3d_18_ucf101/pretraining_for_ucf.py | 34 +++ .../fedvssl/fedvssl/data_partitioning_k400.py | 113 +++++++++ .../fedvssl/fedvssl/data_partitioning_ucf.py | 74 ++++++ baselines/fedvssl/fedvssl/dataset.py | 10 + .../fedvssl/dataset_convert_to_json.py | 29 +++ .../fedvssl/fedvssl/dataset_preparation.py | 4 + .../fedvssl/fedvssl/finetune_preprocess.py | 59 +++++ baselines/fedvssl/fedvssl/main.py | 153 ++++++++++++ baselines/fedvssl/fedvssl/models.py | 7 + baselines/fedvssl/fedvssl/server.py | 5 + baselines/fedvssl/fedvssl/strategy.py | 159 ++++++++++++ baselines/fedvssl/fedvssl/utils.py | 144 +++++++++++ baselines/fedvssl/pyproject.toml | 146 +++++++++++ doc/source/ref-changelog.md | 2 + 28 files changed, 1939 insertions(+) create mode 100644 baselines/fedvssl/.gitignore create mode 100644 baselines/fedvssl/LICENSE create mode 100644 baselines/fedvssl/README.md create mode 100644 baselines/fedvssl/_static/FVSSL.png create mode 100644 baselines/fedvssl/fedvssl/__init__.py create mode 100644 baselines/fedvssl/fedvssl/client.py create mode 100644 baselines/fedvssl/fedvssl/conf/base.yaml create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/finetuning/model_r3d18.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/finetuning/r3d_18_ucf101/finetune_ucf101.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/finetuning/r3d_18_ucf101/test_ucf101.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/finetuning/runtime_ucf101.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/pretraining/pretraining_runtime_kinetics.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/pretraining/pretraining_runtime_ucf.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/pretraining/r3d_18_kinetics/pretraining_for_kinetics.py create mode 100644 baselines/fedvssl/fedvssl/conf/mmcv_conf/pretraining/r3d_18_ucf101/pretraining_for_ucf.py create mode 100644 baselines/fedvssl/fedvssl/data_partitioning_k400.py create mode 100644 baselines/fedvssl/fedvssl/data_partitioning_ucf.py create mode 100644 baselines/fedvssl/fedvssl/dataset.py create mode 100644 baselines/fedvssl/fedvssl/dataset_convert_to_json.py create mode 100644 baselines/fedvssl/fedvssl/dataset_preparation.py create mode 100644 baselines/fedvssl/fedvssl/finetune_preprocess.py create mode 100644 baselines/fedvssl/fedvssl/main.py create mode 100644 baselines/fedvssl/fedvssl/models.py create mode 100644 baselines/fedvssl/fedvssl/server.py create mode 100644 baselines/fedvssl/fedvssl/strategy.py create mode 100644 baselines/fedvssl/fedvssl/utils.py create mode 100644 baselines/fedvssl/pyproject.toml diff --git a/baselines/fedvssl/.gitignore b/baselines/fedvssl/.gitignore new file mode 100644 index 000000000000..a31a594130ab --- /dev/null +++ b/baselines/fedvssl/.gitignore @@ -0,0 +1,2 @@ +outputs/ +*.__pycache__ diff --git a/baselines/fedvssl/LICENSE b/baselines/fedvssl/LICENSE new file mode 100644 index 000000000000..d64569567334 --- /dev/null +++ b/baselines/fedvssl/LICENSE @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/baselines/fedvssl/README.md b/baselines/fedvssl/README.md new file mode 100644 index 000000000000..a9de7a52f54c --- /dev/null +++ b/baselines/fedvssl/README.md @@ -0,0 +1,227 @@ +--- +title: Federated Self-supervised Learning for Video Understanding +url: https://arxiv.org/abs/2207.01975 +labels: [action recognition, cross-device, ssl, video, videossl] +dataset: [UCF-101, Kinectics-400] +--- + +# Federated Self-supervised Learning for Video Understanding +> Note: If you use this baseline in your work, please remember to cite the original authors of the paper as well as the Flower paper. + + +**paper:** [arxiv.org/abs/2207.01975](https://arxiv.org/abs/2207.01975) + + +**Authors:** Yasar Abbas Ur Rehman, Yan Gao, Jiajun Shen, Pedro Porto Buarque de Gusmao, Nicholas Lane + + +**Abstract:** The ubiquity of camera-enabled mobile devices has lead to large amounts of unlabelled video data being produced at the edge. Although various self-supervised learning (SSL) methods have been proposed to harvest their latent spatio-temporal representations for task-specific training, practical challenges including privacy concerns and communication costs prevent SSL from being deployed at large scales. To mitigate these issues, we propose the use of Federated Learning (FL) to the task of video SSL. In this work, we evaluate the performance of current state-of-the-art (SOTA) video-SSL techniques and identify their shortcomings when integrated into the large-scale FL setting simulated with kinetics-400 dataset. We follow by proposing a novel federated SSL framework for video, dubbed FedVSSL, that integrates different aggregation strategies and partial weight updating. Extensive experiments demonstrate the effectiveness and significance of FedVSSL as it outperforms the centralized SOTA for the downstream retrieval task by 6.66% on UCF-101 and 5.13% on HMDB-51. + +![](_static/FVSSL.png) + +## About this baseline + +**Whats's implemented:** The code in this directory replicates the experiments in *Federated Self-supervised Learning for Video Understanding* (Rehman et al., 2022) for UCF-101, which proposed the FedVSSL algorithm. Specifically, it replicates the results for UCF-101 in Table 4 in the paper. +As common SSL training pipline, this code has two parts: SSL pre-training in FL and downstream fine-tuning. + +**Dataset:** [UCF-101](https://www.crcv.ucf.edu/data/UCF101.php) + +**Hardware Setup:** These experiments (SSL pre-train + downstream fine-tuning) were on a server with 6x RTX-3090 GPU and 128 CPU threads. Assuming a single client runs per GPU, on a RTX 3090 it takes 34s for a client to complete its local training (1 epoch); on an A100 this goes down to 17s. In this baseline 5 clients participate in each round for a total of 20 rounds. On a single A100 you can run the federated pre-training in under 1h. + +**Contributers:** Yasar Abbas Ur Rehman and Yan Gao + +## Experimental Setup + +**Task:** Action Recognition + +**Model:** +* We first pre-train the R3D-18 model using the Catch-the-Patch (CtP [paper](https://arxiv.org/abs/2105.02545)) SSL technique in FL settings. The details of the CtP SSL technique can be found at `fedvssl/CtP/pyvrl/models/pretraining/ctp`, and the details of the R3D-18 model can be found at `fedvssl/CtP/pyvrl/models/backbones/r3d.py`. +* After pre-training, we fine-tune the pre-trained R3D-18 model on the UCF-101 dataset. + +**Dataset:** The pre-training in the paper was conducted on Kinectics-400 dataset, and we provide the script to generate the partitions for FL training. +However, the original dataset source as well as the pre-processing scripts they offered are not available currently. +Alternately, one can download this dataset from PyTorch ([torchvision-datasets-Kinetics](https://pytorch.org/vision/stable/generated/torchvision.datasets.Kinetics.html)). + +In this baseline, we demonstrate SSL pre-training and supervised fine-tuning with UCF-101 dataset. +For UCF-101 dataset, one can simply follow the dataset preparation instruction below to download and generate the FL partitions. + +| Dataset | #classes | #partitions | partitioning method | partition settings | +|:--------|:--------:|:-----------:| :---: |:--------------------:| +| UCF101 | 101 | 10 | randomly partitioned | uniform | +| Kinectics-400 | 400 | 100 | randomly partitioned | 8 classes per client | + +**Training Hyperparameters:** The following table shows the main hyperparameters for this baseline with their default value (i.e. the value used if you run `python main.py` directly) + +| Description | Default Value | +|:-------------------|:-----------------------------------:| +| total clients | 5 | +| clients per round | 5 | +| number of rounds | 20 | +| client resources | {'num_cpus': 2.0, 'num_gpus': 1.0 } | +| optimizer | SGD | +| alpha coefficient | 0.9 | +| beta coefficient | 1 | + +## Environment Setup +Please make sure you have installed CUDA 11.7 on your machine +(see [NVIDIA CUDA Toolkit website](https://developer.nvidia.com/cuda-11-7-0-download-archive)). +To construct the Python environment follow these steps (asuming you arelady have `pyenv` and `Poetry` in your system): + +```bash +# Set directory to use python 3.10 (install with `pyenv install ` if you don't have it) +pyenv local 3.10.12 +poetry env use 3.10.12 + +# Install the base Poetry environment +poetry install + +# Activate the environment +poetry shell +``` + +## Running the Experiments +To run FedVSSL with UCF-101 baseline, please ensure you have activated your Poetry environment (execute `poetry shell` from this directory). +Then, download the `CtP` repo and install required packages: + +```bash +# Clone CtP repo +git clone https://github.com/yan-gao-GY/CtP.git fedvssl/CtP + +# Additional packages to decompress the dataset +sudo apt install unrar unzip +``` + +### Dataset preparation +Let's first download UCF-101 dataset and related annotation files: + +```bash +cd fedvssl +mkdir -p data/ucf101/ + +# Downloading +wget https://www.crcv.ucf.edu/data/UCF101/UCF101.rar -O data/ucf101/UCF101.rar --no-check-certificate + +# Unzipping +unrar e data/ucf101/UCF101.rar data/ucf101/UCF101_raw/ + +# Downloading the train/test split +wget https://www.crcv.ucf.edu/data/UCF101/UCF101TrainTestSplits-RecognitionTask.zip -O data/ucf101/UCF101TrainTestSplits-RecognitionTask.zip --no-check-certificate + +# Unzipping +unzip data/ucf101/UCF101TrainTestSplits-RecognitionTask.zip -d data/ucf101/ +``` + +Then, we use the scripts to pre-process the dataset: + +```bash +# Pre-processing the dataset +python CtP/scripts/process_ucf101.py --raw_dir data/ucf101/UCF101_raw/ --ann_dir data/ucf101/ucfTrainTestlist/ --out_dir data/ucf101/ + +# Covert to .json files +python dataset_convert_to_json.py + +# (optional) +rm data/ucf101/UCF101.rar +rm -r data/ucf101/UCF101_raw/ +``` + +Then, we perform data partitioning for FL: + +```bash +python data_partitioning_ucf.py --json_path data/ucf101/annotations --output_path data/ucf101/annotations/client_distribution/ --num_clients 5 + +cd .. +``` + + +### Federated SSL pre-training + +Finally, we can launch the training. To run SSL pretraining using `FedVSSL`: +```bash +# Run federated SSL training with FedVSSL +python -m fedvssl.main # this will run using the default settings. + +# You can override settings directly from the command line +python -m fedvssl.main strategy.mix_coeff=1 rounds=100 # will set hyper-parameter alpha to 1 and the number of rounds to 100 +``` + +To run using `FedAvg`: + +```bash +# This will run FedAvg baseline +# This is done so to match the experimental setup in the paper +python -m fedvssl.main strategy.fedavg=true + +# This config can also be overriden. +``` + +Running any of the above will create a directory structure in the form of `outputs//