Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Convert Scikit-Learn examples (SVM, Kmeans, Linear) to use Job API #2845

Merged
merged 9 commits into from
Aug 27, 2024
Merged

Convert Scikit-Learn examples (SVM, Kmeans, Linear) to use Job API #2845

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
81ac2d5
convert scikit-learn (Linear, SVM and Kmeans) to use Job API
chesterxgchen Aug 25, 2024
7496fb5
format style
chesterxgchen Aug 25, 2024
57453bd
remove duplicate file
chesterxgchen Aug 25, 2024
c0ad9c6
update based on comments
chesterxgchen Aug 26, 2024
01e1f27
rename executor to runner
chesterxgchen Aug 26, 2024
f9ed2b6
fix typo
chesterxgchen Aug 26, 2024
ba9fbac
tweaks
chesterxgchen Aug 26, 2024
a1fb5f5
Merge branch 'main' into scikit_learn
chesterxgchen Aug 26, 2024
c9590ba
Merge branch 'main' into scikit_learn
chesterxgchen Aug 27, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
63 changes: 63 additions & 0 deletions examples/hello-world/step-by-step/higgs/sklearn-kmeans/code/sklearn_kmeans_job.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
# Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
#
# 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.
from kmeans_assembler import KMeansAssembler

from nvflare import FedJob
from nvflare.app_common.aggregators.collect_and_assemble_aggregator import CollectAndAssembleAggregator
from nvflare.app_common.shareablegenerators import FullModelShareableGenerator
from nvflare.app_common.workflows.scatter_and_gather import ScatterAndGather
from nvflare.app_opt.sklearn.joblib_model_param_persistor import JoblibModelParamPersistor
from nvflare.job_config.script_runner import FrameworkType, ScriptRunner

if __name__ == "__main__":
n_clients = 3
num_rounds = 20
train_script = "code/kmeans_fl.py"
script_args = "--data_root_dir /tmp/nvflare/dataset/output"

aggregator_id = "aggregator"
persistor_id = "persistor"
shareable_generator_id = "shareable_generator"
assembler_id = "kmeans_assembler"

job = FedJob("sklearn_kmeans")

initial_params = dict(n_clusters=2)
job.to(JoblibModelParamPersistor(initial_params=initial_params), "server", id=persistor_id)
job.to(FullModelShareableGenerator(), "server", id=shareable_generator_id)
job.to(CollectAndAssembleAggregator(assembler_id=assembler_id), "server", id=aggregator_id)
job.to(KMeansAssembler(), "server", id=assembler_id)

ctrl = ScatterAndGather(
min_clients=n_clients,
num_rounds=num_rounds,
start_round=0,
wait_time_after_min_received=0,
aggregator_id=aggregator_id,
persistor_id=persistor_id,
shareable_generator_id=shareable_generator_id,
train_task_name="train",
train_timeout=0,
allow_empty_global_weights=True,
)

job.to(ctrl, "server")

# Add clients
for i in range(n_clients):
runner = ScriptRunner(script=train_script, script_args=script_args, framework=FrameworkType.RAW)
job.to(runner, f"site-{i + 1}")

job.export_job("/tmp/nvflare/jobs")
job.simulator_run("/tmp/nvflare/sklearn_kmeans", gpu="0")
98 changes: 9 additions & 89 deletions examples/hello-world/step-by-step/higgs/sklearn-kmeans/sklearn_kmeans.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,11 +21,8 @@
"\n",
"## Setup NVFLARE\n",
"\n",
"Follow [Getting Started](https://nvflare.readthedocs.io/en/main/getting_started.html) to set up a virtual environment and install NVFLARE.\n",
"\n",
"You can also follow this [notebook](https://github.com/NVIDIA/NVFlare/blob/main/examples/nvflare_setup.ipynb) to get set up.\n",
"\n",
"> Make sure you have installed nvflare from **terminal** \n"
"Follow [Getting Started](../../../../getting_started/readme.ipynb) to set up a virtual environment and install NVFLARE.\n",
"\n"
]
},
{
Expand Down Expand Up @@ -314,77 +311,8 @@
"\n",
"## Prepare Job \n",
"\n",
"Now, we have the code, we need to prepare job folder with configurations to run in NVFLARE. To do this, we can leveage the job template for scikit learn. First look at the the available job templates"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "01fdd962-e63c-4b58-81e7-beeedd05509b",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare config -jt ../../../../../job_templates/"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a5ed150d-4692-49fe-87a6-1779ec64d9df",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare job list_templates"
]
},
{
"cell_type": "markdown",
"id": "d7602cec-5fb5-4b23-a82a-b7444f2af471",
"metadata": {},
"source": [
"the `sklearn_kmeans` is the one we need. "
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "9899a684-444f-4c04-a388-4bdd9b7dc649",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare job create -j /tmp/nvflare/jobs/sklearn_kmeans -force -w sklearn_kmeans \\\n",
"-sd code \\\n",
"-f config_fed_client.conf app_script=\"kmeans_fl.py\" app_config=\"--data_root_dir /tmp/nvflare/dataset/output\""
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "17dc5dcc-573e-479e-965d-0cca8b58995a",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!cat /tmp/nvflare/jobs/sklearn_kmeans/app/config/config_fed_client.conf"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7b3086fd-ecf3-4584-a643-5706ae2069b5",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!tree /tmp/nvflare/jobs/sklearn_kmeans"
"Now, we have the code, we need to prepare job folder with configurations to run in NVFLARE. To do this, we are going to use Job API to construct a FedJob, which can be used to run simulation and export Job config. \n",
"\n"
]
},
{
Expand All @@ -410,13 +338,13 @@
{
"cell_type": "code",
"execution_count": null,
"id": "09800691-46e0-4a95-bea6-d2a4a425052b",
"id": "dcb89b46-aad1-433d-b2b1-6b46380cef7c",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare simulator /tmp/nvflare/jobs/sklearn_kmeans -w /tmp/nvflare/sklearn_kmeans -n 3 -t 3"
"!python3 code/sklearn_kmeans_job.py"
]
},
{
Expand All @@ -428,7 +356,7 @@
"source": [
"Let's examine the results.\n",
"\n",
"We can notice from the FL training log, at the last round of local training, site-1 reports `site-1: global model homogeneity_score: 0.0068`\n",
"We can notice from the FL training log, at the last round of local training, site-1 reports `site-1: global model homogeneity_score: 0.0040`\n",
"Now let's run a local training to verify if this number makes sense."
]
},
Expand All @@ -449,22 +377,14 @@
"id": "ea7bbacc-b059-4f82-9785-2b22bf840ef9",
"metadata": {},
"source": [
"HIGGS dataset is challenging for unsupervised clustering, as we can observe from the result. As shown by the local training with same number of iterations, the score is `model homogeneity_score: 0.0049`. As compared with the FL score of `0.0068`, FL in this case still provides some benefit from the collaborative learning.\n",
"HIGGS dataset is challenging for unsupervised clustering, as we can observe from the result. As shown by the local training with same number of iterations, the score is `model homogeneity_score: 0.0037`. As compared with the FL score of `0.0040`, FL in this case still provides some benefit from the collaborative learning.\n",
"\n",
"## We are done !\n",
"Congratulations! you have just completed the federated k-Means clustering for tabular data. \n",
"\n",
"Now we will move on from scikit-learn and take a look at how to use federated XGBoost.\n",
"In the next example [xgboost](../xgboost/xgboost_horizontal.ipynb), we will show a federated horizontal xgboost learning with bagging collaboration."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "b5329a14-1a81-488a-8edc-2b7c85cffd8c",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
Expand All @@ -483,7 +403,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.16"
"version": "3.8.19"
}
},
"nbformat": 4,
Expand Down
89 changes: 21 additions & 68 deletions examples/hello-world/step-by-step/higgs/sklearn-linear/sklearn_linear.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,11 +21,7 @@
"\n",
"## Setup NVFLARE\n",
"\n",
"Follow [Getting Started](https://nvflare.readthedocs.io/en/main/getting_started.html) to set up a virtual environment and install NVFLARE.\n",
"\n",
"You can also follow this [notebook](https://github.com/NVIDIA/NVFlare/blob/main/examples/nvflare_setup.ipynb) to get set up.\n",
"\n",
"> Make sure you have installed nvflare from **terminal** \n"
"Follow [Getting Started](../../../../getting_started/readme.ipynb) to set up a virtual environment and install NVFLARE.\n"
]
},
{
Expand Down Expand Up @@ -307,7 +303,7 @@
"\n",
"## Prepare Job \n",
"\n",
"Now, we have the code, we need to prepare job folder with configurations to run in NVFLARE. To do this, we can leveage the job template for scikit learn. First look at the the available job templates"
"Now, we have the code, we need to prepare job folder with configurations to run in NVFLARE. To do this, we are going to use Job API to construct a FedJob, which can be used to run simulation and export Job config. "
]
},
{
Expand All @@ -319,65 +315,7 @@
},
"outputs": [],
"source": [
"!nvflare config -jt ../../../../../job_templates/"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a5ed150d-4692-49fe-87a6-1779ec64d9df",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare job list_templates"
]
},
{
"cell_type": "markdown",
"id": "d7602cec-5fb5-4b23-a82a-b7444f2af471",
"metadata": {},
"source": [
"the `sklearn_linear` is the one we need. "
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "9899a684-444f-4c04-a388-4bdd9b7dc649",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare job create -j /tmp/nvflare/jobs/sklearn_sgd -force -w sklearn_linear \\\n",
"-sd code \\\n",
"-f config_fed_client.conf app_script=\"sgd_fl.py\" app_config=\"--data_root_dir /tmp/nvflare/dataset/output\"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "17dc5dcc-573e-479e-965d-0cca8b58995a",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!cat /tmp/nvflare/jobs/sklearn_sgd/app/config/config_fed_client.conf"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7b3086fd-ecf3-4584-a643-5706ae2069b5",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!tree /tmp/nvflare/jobs/sklearn_sgd"
"!cat sklearn_linear_job.py"
]
},
{
Expand All @@ -403,13 +341,28 @@
{
"cell_type": "code",
"execution_count": null,
"id": "09800691-46e0-4a95-bea6-d2a4a425052b",
"id": "6a068316-b40b-407a-b886-b5c1c3082801",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!nvflare simulator /tmp/nvflare/jobs/sklearn_sgd -w /tmp/nvflare/sklearn_sgd -n 3 -t 3"
"!python3 sklearn_linear_job.py"
]
},
{
"cell_type": "markdown",
"id": "dad05c48-b60d-4bcd-a13f-6b52056a804f",
"metadata": {},
"source": [
"You also run from the exported job folder using **Simulator CLI**\n",
"\n",
"``` \n",
"\n",
"!nvflare simulator /tmp/nvflare/jobs/sklearn_sgd -w /tmp/nvflare/sklearn_sgd -n 3 -t 3\n",
"\n",
"\n",
"```"
]
},
{
Expand Down Expand Up @@ -489,7 +442,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.16"
"version": "3.8.19"
}
},
"nbformat": 4,
Expand Down
Loading
Loading