Hyperparameter Optimization in AutoMM#

Hyperparameter optimization (HPO) is a method that helps solve the challenge of tuning hyperparameters of machine learning models. ML algorithms have multiple complex hyperparameters that generate an enormous search space, and the search space in deep learning methods is even larger than traditional ML algorithms. Tuning on a massive search space is a tough challenge, but AutoMM provides various options for you to guide the fitting process based on your domain knowledge and the constraint on computing resources.

Create Image Dataset#

In this tutorial, we are going to again use the subset of the Shopee-IET dataset from Kaggle for demonstration purpose. Each image contains a clothing item and the corresponding label specifies its clothing category. Our subset of the data contains the following possible labels: BabyPants, BabyShirt, womencasualshoes, womenchiffontop.

We can load a dataset by downloading a url data automatically:

import warnings
warnings.filterwarnings('ignore')
from datetime import datetime

from autogluon.multimodal.utils.misc import shopee_dataset
download_dir = './ag_automm_tutorial_hpo'
train_data, test_data = shopee_dataset(download_dir)
train_data = train_data.sample(frac=0.5)
print(train_data)

Downloading ./ag_automm_tutorial_hpo/file.zip from https://automl-mm-bench.s3.amazonaws.com/vision_datasets/shopee.zip...
                                                 image  label
/home/ci/autogluon/docs/tutorials/multimodal/a...      2
/home/ci/autogluon/docs/tutorials/multimodal/a...      0
  /home/ci/autogluon/docs/tutorials/multimodal/a...      0
/home/ci/autogluon/docs/tutorials/multimodal/a...      1
/home/ci/autogluon/docs/tutorials/multimodal/a...      2
..                                                 ...    ...
/home/ci/autogluon/docs/tutorials/multimodal/a...      1
/home/ci/autogluon/docs/tutorials/multimodal/a...      2
/home/ci/autogluon/docs/tutorials/multimodal/a...      0
/home/ci/autogluon/docs/tutorials/multimodal/a...      1
/home/ci/autogluon/docs/tutorials/multimodal/a...      1

[400 rows x 2 columns]

100%|██████████| 41.9M/41.9M [00:00<00:00, 87.3MiB/s]

There are in total 400 data points in this dataset. The image column stores the path to the actual image, and the label column stands for the label class.

The Regular Model Fitting#

Recall that if we are to use the default settings predefined by Autogluon, we can simply fit the model using MultiModalPredictor with three lines of code:

from autogluon.multimodal import MultiModalPredictor
predictor_regular = MultiModalPredictor(label="label")
start_time = datetime.now()
predictor_regular.fit(
    train_data=train_data,
    hyperparameters = {"model.timm_image.checkpoint_name": "ghostnet_100"}
)
end_time = datetime.now()
elapsed_seconds = (end_time - start_time).total_seconds()
elapsed_min = divmod(elapsed_seconds, 60)
print("Total fitting time: ", f"{int(elapsed_min[0])}m{int(elapsed_min[1])}s")

Total fitting time:  0m47s

No path specified. Models will be saved in: "AutogluonModels/ag-20230629_231640/"
AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == int, but few unique label-values observed).
	4 unique label values:  [1, 2, 0, 3]
	If 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression'])
Global seed set to 0
AutoMM starts to create your model. ✨

- AutoGluon version is 0.8.1b20230629.

- Pytorch version is 1.13.1+cu117.

- Model will be saved to "/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640".

- Validation metric is "accuracy".

- To track the learning progress, you can open a terminal and launch Tensorboard:
    ```shell
    # Assume you have installed tensorboard
    tensorboard --logdir /home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640
    ```

Enjoy your coffee, and let AutoMM do the job ☕☕☕ Learn more at https://auto.gluon.ai

Downloading: "https://github.com/huawei-noah/CV-backbones/releases/download/ghostnet_pth/ghostnet_1x.pth" to /home/ci/.cache/torch/hub/checkpoints/ghostnet_1x.pth
1 GPUs are detected, and 1 GPUs will be used.
   - GPU 0 name: Tesla T4
   - GPU 0 memory: 15.74GB/15.84GB (Free/Total)
CUDA version is 11.7.

Using 16bit None Automatic Mixed Precision (AMP)
GPU available: True (cuda), used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]

  | Name              | Type                            | Params
----------------------------------------------------------------------
0 | model             | TimmAutoModelForImagePrediction | 3.9 M 
1 | validation_metric | MulticlassAccuracy              | 0     
2 | loss_func         | CrossEntropyLoss                | 0     
----------------------------------------------------------------------
3.9 M     Trainable params
0         Non-trainable params
3.9 M     Total params
7.813     Total estimated model params size (MB)
Epoch 0, global step 1: 'val_accuracy' reached 0.21250 (best 0.21250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=0-step=1.ckpt' as top 3
Epoch 0, global step 3: 'val_accuracy' reached 0.26250 (best 0.26250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=0-step=3.ckpt' as top 3
Epoch 1, global step 4: 'val_accuracy' reached 0.28750 (best 0.28750), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=1-step=4.ckpt' as top 3
Epoch 1, global step 6: 'val_accuracy' reached 0.41250 (best 0.41250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=1-step=6.ckpt' as top 3
Epoch 2, global step 7: 'val_accuracy' reached 0.40000 (best 0.41250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=2-step=7.ckpt' as top 3
Epoch 2, global step 9: 'val_accuracy' reached 0.50000 (best 0.50000), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=2-step=9.ckpt' as top 3
Epoch 3, global step 10: 'val_accuracy' reached 0.48750 (best 0.50000), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=3-step=10.ckpt' as top 3
Epoch 3, global step 12: 'val_accuracy' reached 0.55000 (best 0.55000), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=3-step=12.ckpt' as top 3
Epoch 4, global step 13: 'val_accuracy' reached 0.56250 (best 0.56250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=4-step=13.ckpt' as top 3
Epoch 4, global step 15: 'val_accuracy' reached 0.61250 (best 0.61250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=4-step=15.ckpt' as top 3
Epoch 5, global step 16: 'val_accuracy' reached 0.61250 (best 0.61250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=5-step=16.ckpt' as top 3
Epoch 5, global step 18: 'val_accuracy' reached 0.61250 (best 0.61250), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=5-step=18.ckpt' as top 3
Epoch 6, global step 19: 'val_accuracy' was not in top 3
Epoch 6, global step 21: 'val_accuracy' reached 0.62500 (best 0.62500), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=6-step=21.ckpt' as top 3
Epoch 7, global step 22: 'val_accuracy' reached 0.63750 (best 0.63750), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=7-step=22.ckpt' as top 3
Epoch 7, global step 24: 'val_accuracy' was not in top 3
Epoch 8, global step 25: 'val_accuracy' reached 0.63750 (best 0.63750), saving model to '/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640/epoch=8-step=25.ckpt' as top 3
Epoch 8, global step 27: 'val_accuracy' was not in top 3
Epoch 9, global step 28: 'val_accuracy' was not in top 3
Epoch 9, global step 30: 'val_accuracy' was not in top 3
`Trainer.fit` stopped: `max_epochs=10` reached.
Start to fuse 3 checkpoints via the greedy soup algorithm.
AutoMM has created your model 🎉🎉🎉

- To load the model, use the code below:
    ```python
    from autogluon.multimodal import MultiModalPredictor
    predictor = MultiModalPredictor.load("/home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640")
    ```

- You can open a terminal and launch Tensorboard to visualize the training log:
    ```shell
    # Assume you have installed tensorboard
    tensorboard --logdir /home/ci/autogluon/docs/tutorials/multimodal/advanced_topics/AutogluonModels/ag-20230629_231640
    ```

- If you are not satisfied with the model, try to increase the training time, 
adjust the hyperparameters (https://auto.gluon.ai/stable/tutorials/multimodal/advanced_topics/customization.html),
or post issues on GitHub: https://github.com/autogluon/autogluon

Let’s check out the test accuracy of the fitted model:

scores = predictor_regular.evaluate(test_data, metrics=["accuracy"])
print('Top-1 test acc: %.3f' % scores["accuracy"])

Top-1 test acc: 0.625

Use HPO During Model Fitting#

If you would like more control over the fitting process, you can specify various options for hyperparameter optimizations(HPO) in MultiModalPredictor by simply adding more options in hyperparameter and hyperparameter_tune_kwargs.

There are a few options we can have in MultiModalPredictor. We use Ray Tune tune library in the backend, so we need to pass in a Tune search space or an AutoGluon search space which will be converted to Tune search space.

Defining the search space of various hyperparameter values for the training of neural networks:

hyperparameters = {
        "optimization.learning_rate": tune.uniform(0.00005, 0.005),
        "optimization.optim_type": tune.choice(["adamw", "sgd"]),
        "optimization.max_epochs": tune.choice(["10", "20"]), 
        "model.timm_image.checkpoint_name": tune.choice(["swin_base_patch4_window7_224", "convnext_base_in22ft1k"])
        }

This is an example but not an exhaustive list. You can find the full supported list in Customize AutoMM

Defining the search strategy for HPO with hyperparameter_tune_kwargs. You can pass in a string or initialize a ray.tune.schedulers.TrialScheduler object.

"searcher": "bayes"

"scheduler": "ASHA"

"num_trials": 20

Let’s work on HPO with combinations of different learning rates and backbone models:

from ray import tune

predictor_hpo = MultiModalPredictor(label="label")

hyperparameters = {
            "optimization.learning_rate": tune.uniform(0.00005, 0.001),
            "model.timm_image.checkpoint_name": tune.choice(["ghostnet_100",
                                                             "mobilenetv3_large_100"])
}
hyperparameter_tune_kwargs = {
    "searcher": "bayes", # random
    "scheduler": "ASHA",
    "num_trials": 2,
}
start_time_hpo = datetime.now()
predictor_hpo.fit(
        train_data=train_data,
        hyperparameters=hyperparameters,
        hyperparameter_tune_kwargs=hyperparameter_tune_kwargs,
    )
end_time_hpo = datetime.now()
elapsed_seconds_hpo = (end_time_hpo - start_time_hpo).total_seconds()
elapsed_min_hpo = divmod(elapsed_seconds_hpo, 60)
print("Total fitting time: ", f"{int(elapsed_min_hpo[0])}m{int(elapsed_min_hpo[1])}s")

No path specified. Models will be saved in: "AutogluonModels/ag-20230629_231729/"
AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == int, but few unique label-values observed).
	4 unique label values:  [1, 2, 0, 3]
	If 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression'])
/home/ci/opt/venv/lib/python3.8/site-packages/ray/tune/trainable/function_trainable.py:612: DeprecationWarning: `checkpoint_dir` in `func(config, checkpoint_dir)` is being deprecated. To save and load checkpoint in trainable functions, please use the `ray.air.session` API:

from ray.air import session

def train(config):
    # ...
    session.report({"metric": metric}, checkpoint=checkpoint)

For more information please see https://docs.ray.io/en/master/tune/api_docs/trainable.html

  warnings.warn(
Removing non-optimal trials and only keep the best one.
Start to fuse 3 checkpoints via the greedy soup algorithm.

Tune Status

Current time:	2023-06-29 23:19:10
Running for:	00:01:36.36
Memory:	7.6/31.0 GiB

System Info

Using AsyncHyperBand: num_stopped=0
Bracket: Iter 4096.000: None | Iter 1024.000: None | Iter 256.000: None | Iter 64.000: None | Iter 16.000: 0.8406250029802322 | Iter 4.000: 0.6562499850988388 | Iter 1.000: 0.27812501043081284
Resources requested: 0/8 CPUs, 0/1 GPUs, 0.0/12.04 GiB heap, 0.0/6.02 GiB objects (0.0/1.0 accelerator_type:T4)

Trial Status

Trial name	status	loc	model.names	model.timm_image.che ckpoint_name	optimization.learnin g_rate	iter	total time (s)	val_accuracy
398c97a5	TERMINATED	10.0.0.89:3165	('categorical_m_d780	ghostnet_100	0.000161821	20	44.5423	0.725
80a37630	TERMINATED	10.0.0.89:3332	('categorical_m_f080	mobilenetv3_lar_a1c0	0.000538089	20	36.6078	0.875

Trial Progress

Trial name	should_checkpoint	val_accuracy
398c97a5	True	0.725
80a37630	True	0.875

Total fitting time:  1m44s

Let’s check out the test accuracy of the fitted model after HPO:

scores_hpo = predictor_hpo.evaluate(test_data, metrics=["accuracy"])
print('Top-1 test acc: %.3f' % scores_hpo["accuracy"])

Top-1 test acc: 0.825

From the training log, you should be able to see the current best trial as below:

Current best trial: 47aef96a with val_accuracy=0.862500011920929 and parameters={'optimization.learning_rate': 0.0007195214018085505, 'model.timm_image.checkpoint_name': 'ghostnet_100'}

After our simple 2-trial HPO run, we got a better test accuracy, by searching different learning rates and models, compared to the out-of-box solution provided in the previous section. HPO helps select the combination of hyperparameters with highest validation accuracy.

Other Examples#

You may go to AutoMM Examples to explore other examples about AutoMM.

Customization#

To learn how to customize AutoMM, please refer to Customize AutoMM.