Global Wheat Challenge 2021
Submit with WILDS
WILDS is a benchmark of in-the-wild distribution shifts spanning diverse data modalities
WILDS is a benchmark of in-the-wild distribution shifts spanning diverse data modalities and applications, from tumor identification to wildlife monitoring to poverty mapping. It focuses on training robust algorithm. You can try to get you first model trained with the toolbox following the notebook !
Global Wheat Competition 2021 - Starting notebook using the WILDS library¶
- WILDS (https://github.com/p-lambda/wilds) is a PyTorch-based library for distribution shifts. It contains data loaders and evaluation functions for the Global Wheat Competition, as well as some examples of algorithms and models that you can build on.
- The goal of this notebook is to help you get started with using the WILDS library to train and submit your first model!
- Using WILDS is not necessary for participating in the competition. However, you might find a lot of the utilities and infrastructure in the WILDS library useful.
- Before starting, please check in Edit / Notebook settings that "GPU" is selected
Download Aicrowd-cli 📚¶
The Aicrowd CLI enables making submissions directly via the notebook.
In [ ]:
!pip install aicrowd-cli
Collecting aicrowd-cli
Downloading https://files.pythonhosted.org/packages/a5/8a/fca67e8c1cb1501a9653cd653232bf6fdebbb2393e3de861aad3636a1136/aicrowd_cli-0.1.6-py3-none-any.whl (51kB)
|████████████████████████████████| 61kB 8.6MB/s
Collecting rich<11,>=10.0.0
Downloading https://files.pythonhosted.org/packages/7b/ff/23056b2b3dfb22bd1400b49639f00945ec320647700c0b288dc29d4ed70a/rich-10.2.0-py3-none-any.whl (203kB)
|████████████████████████████████| 204kB 16.5MB/s
Collecting gitpython<4,>=3.1.12
Downloading https://files.pythonhosted.org/packages/27/da/6f6224fdfc47dab57881fe20c0d1bc3122be290198ba0bf26a953a045d92/GitPython-3.1.17-py3-none-any.whl (166kB)
|████████████████████████████████| 174kB 49.1MB/s
Requirement already satisfied: toml<1,>=0.10.2 in /usr/local/lib/python3.7/dist-packages (from aicrowd-cli) (0.10.2)
Collecting tqdm<5,>=4.56.0
Downloading https://files.pythonhosted.org/packages/72/8a/34efae5cf9924328a8f34eeb2fdaae14c011462d9f0e3fcded48e1266d1c/tqdm-4.60.0-py2.py3-none-any.whl (75kB)
|████████████████████████████████| 81kB 11.0MB/s
Collecting requests-toolbelt<1,>=0.9.1
Downloading https://files.pythonhosted.org/packages/60/ef/7681134338fc097acef8d9b2f8abe0458e4d87559c689a8c306d0957ece5/requests_toolbelt-0.9.1-py2.py3-none-any.whl (54kB)
|████████████████████████████████| 61kB 8.1MB/s
Collecting click<8,>=7.1.2
Downloading https://files.pythonhosted.org/packages/d2/3d/fa76db83bf75c4f8d338c2fd15c8d33fdd7ad23a9b5e57eb6c5de26b430e/click-7.1.2-py2.py3-none-any.whl (82kB)
|████████████████████████████████| 92kB 11.3MB/s
Collecting requests<3,>=2.25.1
Downloading https://files.pythonhosted.org/packages/29/c1/24814557f1d22c56d50280771a17307e6bf87b70727d975fd6b2ce6b014a/requests-2.25.1-py2.py3-none-any.whl (61kB)
|████████████████████████████████| 61kB 9.2MB/s
Collecting colorama<0.5.0,>=0.4.0
Downloading https://files.pythonhosted.org/packages/44/98/5b86278fbbf250d239ae0ecb724f8572af1c91f4a11edf4d36a206189440/colorama-0.4.4-py2.py3-none-any.whl
Requirement already satisfied: typing-extensions<4.0.0,>=3.7.4; python_version < "3.8" in /usr/local/lib/python3.7/dist-packages (from rich<11,>=10.0.0->aicrowd-cli) (3.7.4.3)
Collecting commonmark<0.10.0,>=0.9.0
Downloading https://files.pythonhosted.org/packages/b1/92/dfd892312d822f36c55366118b95d914e5f16de11044a27cf10a7d71bbbf/commonmark-0.9.1-py2.py3-none-any.whl (51kB)
|████████████████████████████████| 51kB 7.5MB/s
Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /usr/local/lib/python3.7/dist-packages (from rich<11,>=10.0.0->aicrowd-cli) (2.6.1)
Collecting gitdb<5,>=4.0.1
Downloading https://files.pythonhosted.org/packages/ea/e8/f414d1a4f0bbc668ed441f74f44c116d9816833a48bf81d22b697090dba8/gitdb-4.0.7-py3-none-any.whl (63kB)
|████████████████████████████████| 71kB 10.6MB/s
Requirement already satisfied: chardet<5,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.25.1->aicrowd-cli) (3.0.4)
Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.25.1->aicrowd-cli) (2.10)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.25.1->aicrowd-cli) (1.24.3)
Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.25.1->aicrowd-cli) (2020.12.5)
Collecting smmap<5,>=3.0.1
Downloading https://files.pythonhosted.org/packages/68/ee/d540eb5e5996eb81c26ceffac6ee49041d473bc5125f2aa995cf51ec1cf1/smmap-4.0.0-py2.py3-none-any.whl
ERROR: google-colab 1.0.0 has requirement requests~=2.23.0, but you'll have requests 2.25.1 which is incompatible.
ERROR: datascience 0.10.6 has requirement folium==0.2.1, but you'll have folium 0.8.3 which is incompatible.
Installing collected packages: colorama, commonmark, rich, smmap, gitdb, gitpython, tqdm, requests, requests-toolbelt, click, aicrowd-cli
Found existing installation: tqdm 4.41.1
Uninstalling tqdm-4.41.1:
Successfully uninstalled tqdm-4.41.1
Found existing installation: requests 2.23.0
Uninstalling requests-2.23.0:
Successfully uninstalled requests-2.23.0
Found existing installation: click 8.0.0
Uninstalling click-8.0.0:
Successfully uninstalled click-8.0.0
Successfully installed aicrowd-cli-0.1.6 click-7.1.2 colorama-0.4.4 commonmark-0.9.1 gitdb-4.0.7 gitpython-3.1.17 requests-2.25.1 requests-toolbelt-0.9.1 rich-10.2.0 smmap-4.0.0 tqdm-4.60.0
In [ ]:
### Please enter your API Key from [here](https://www.aicrowd.com/participants/me).
API_KEY = ""
!aicrowd login --api-key $API_KEY
API Key valid Saved API Key successfully!
Download the WILDS library¶
For more detailed installation instructions, please refer to the WILDS documentation.
In [ ]:
!pip uninstall torch torchvision torchaudio -y
!pip install torch==1.7.1+cu101 torchvision==0.8.2+cu101 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html
!pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.7.0+cu101.html
!git clone https://github.com/p-lambda/wilds/
!cd wilds && git checkout dev && pip install -e .
!pip install transformers
Uninstalling torch-1.8.1+cu101: Successfully uninstalled torch-1.8.1+cu101 Uninstalling torchvision-0.9.1+cu101: Successfully uninstalled torchvision-0.9.1+cu101 WARNING: Skipping torchaudio as it is not installed. Looking in links: https://download.pytorch.org/whl/torch_stable.html Collecting torch==1.7.1+cu101 Downloading https://download.pytorch.org/whl/cu101/torch-1.7.1%2Bcu101-cp37-cp37m-linux_x86_64.whl (735.4MB) |████████████████████████████████| 735.4MB 24kB/s Collecting torchvision==0.8.2+cu101 Downloading https://download.pytorch.org/whl/cu101/torchvision-0.8.2%2Bcu101-cp37-cp37m-linux_x86_64.whl (12.8MB) |████████████████████████████████| 12.8MB 264kB/s Collecting torchaudio==0.7.2 Downloading https://files.pythonhosted.org/packages/37/16/ecdb9eb09ec6b8133d6c9536ea9e49cd13c9b5873c8488b8b765a39028da/torchaudio-0.7.2-cp37-cp37m-manylinux1_x86_64.whl (7.6MB) |████████████████████████████████| 7.6MB 17.8MB/s Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch==1.7.1+cu101) (3.7.4.3) Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from torch==1.7.1+cu101) (1.19.5) Requirement already satisfied: pillow>=4.1.1 in /usr/local/lib/python3.7/dist-packages (from torchvision==0.8.2+cu101) (7.1.2) ERROR: torchtext 0.9.1 has requirement torch==1.8.1, but you'll have torch 1.7.1+cu101 which is incompatible. Installing collected packages: torch, torchvision, torchaudio Successfully installed torch-1.7.1+cu101 torchaudio-0.7.2 torchvision-0.8.2+cu101 Looking in links: https://pytorch-geometric.com/whl/torch-1.7.0+cu101.html Collecting torch-scatter Downloading https://pytorch-geometric.com/whl/torch-1.7.0%2Bcu101/torch_scatter-2.0.6-cp37-cp37m-linux_x86_64.whl (2.8MB) |████████████████████████████████| 2.8MB 11.9MB/s Installing collected packages: torch-scatter Successfully installed torch-scatter-2.0.6 Cloning into 'wilds'... remote: Enumerating objects: 2835, done. remote: Counting objects: 100% (517/517), done. remote: Compressing objects: 100% (119/119), done. remote: Total 2835 (delta 437), reused 435 (delta 396), pack-reused 2318 Receiving objects: 100% (2835/2835), 695.00 KiB | 23.17 MiB/s, done. Resolving deltas: 100% (2114/2114), done. Branch 'dev' set up to track remote branch 'dev' from 'origin'. Switched to a new branch 'dev' Obtaining file:///content/wilds Requirement already satisfied: numpy>=1.19.1 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (1.19.5) Requirement already satisfied: pandas>=1.1.0 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (1.1.5) Requirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (0.22.2.post1) Collecting pillow>=7.2.0 Downloading https://files.pythonhosted.org/packages/33/34/542152297dcc6c47a9dcb0685eac6d652d878ed3cea83bf2b23cb988e857/Pillow-8.2.0-cp37-cp37m-manylinux1_x86_64.whl (3.0MB) |████████████████████████████████| 3.0MB 26.7MB/s Requirement already satisfied: torch>=1.7.0 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (1.7.1+cu101) Collecting ogb>=1.2.6 Downloading https://files.pythonhosted.org/packages/d2/c5/20b1e4a5ff90ead06139ce1c2362474b97bb3a73ee0166eb37f2d3eb0dba/ogb-1.3.1-py3-none-any.whl (67kB) |████████████████████████████████| 71kB 11.1MB/s Requirement already satisfied: tqdm>=4.53.0 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (4.60.0) Collecting outdated>=0.2.0 Downloading https://files.pythonhosted.org/packages/fd/f6/95588d496e518355c33b389222c99069b1c6f2c046be64f400072fdc7cda/outdated-0.2.1-py3-none-any.whl Collecting pytz>=2020.4 Downloading https://files.pythonhosted.org/packages/70/94/784178ca5dd892a98f113cdd923372024dc04b8d40abe77ca76b5fb90ca6/pytz-2021.1-py2.py3-none-any.whl (510kB) |████████████████████████████████| 512kB 49.7MB/s Requirement already satisfied: torchvision==0.8.2 in /usr/local/lib/python3.7/dist-packages (from wilds==1.1.0) (0.8.2+cu101) Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=1.1.0->wilds==1.1.0) (2.8.1) Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.20.0->wilds==1.1.0) (1.0.1) Requirement already satisfied: scipy>=0.17.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.20.0->wilds==1.1.0) (1.4.1) Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch>=1.7.0->wilds==1.1.0) (3.7.4.3) Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.7/dist-packages (from ogb>=1.2.6->wilds==1.1.0) (1.15.0) Requirement already satisfied: urllib3>=1.24.0 in /usr/local/lib/python3.7/dist-packages (from ogb>=1.2.6->wilds==1.1.0) (1.24.3) Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from outdated>=0.2.0->wilds==1.1.0) (2.25.1) Collecting littleutils Downloading https://files.pythonhosted.org/packages/4e/b1/bb4e06f010947d67349f863b6a2ad71577f85590180a935f60543f622652/littleutils-0.2.2.tar.gz Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->outdated>=0.2.0->wilds==1.1.0) (2020.12.5) Requirement already satisfied: chardet<5,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->outdated>=0.2.0->wilds==1.1.0) (3.0.4) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->outdated>=0.2.0->wilds==1.1.0) (2.10) Building wheels for collected packages: littleutils Building wheel for littleutils (setup.py) ... done Created wheel for littleutils: filename=littleutils-0.2.2-cp37-none-any.whl size=7051 sha256=35df43dd0cd025e7469306166212eecced8f95726f2f9f72ce484c2939f7de59 Stored in directory: /root/.cache/pip/wheels/53/16/9f/ac67d15c40243754fd73f620e1b9b6dedc20492ecc19a2bae1 Successfully built littleutils ERROR: google-colab 1.0.0 has requirement requests~=2.23.0, but you'll have requests 2.25.1 which is incompatible. ERROR: albumentations 0.1.12 has requirement imgaug<0.2.7,>=0.2.5, but you'll have imgaug 0.2.9 which is incompatible. Installing collected packages: pillow, littleutils, outdated, ogb, pytz, wilds Found existing installation: Pillow 7.1.2 Uninstalling Pillow-7.1.2: Successfully uninstalled Pillow-7.1.2 Found existing installation: pytz 2018.9 Uninstalling pytz-2018.9: Successfully uninstalled pytz-2018.9 Running setup.py develop for wilds Successfully installed littleutils-0.2.2 ogb-1.3.1 outdated-0.2.1 pillow-8.2.0 pytz-2021.1 wilds Collecting transformers Downloading https://files.pythonhosted.org/packages/b0/9e/5b80becd952d5f7250eaf8fc64b957077b12ccfe73e9c03d37146ab29712/transformers-4.6.0-py3-none-any.whl (2.3MB) |████████████████████████████████| 2.3MB 30.5MB/s Requirement already satisfied: importlib-metadata; python_version < "3.8" in /usr/local/lib/python3.7/dist-packages (from transformers) (4.0.1) Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.19.5) Collecting sacremoses Downloading https://files.pythonhosted.org/packages/75/ee/67241dc87f266093c533a2d4d3d69438e57d7a90abb216fa076e7d475d4a/sacremoses-0.0.45-py3-none-any.whl (895kB) |████████████████████████████████| 901kB 45.7MB/s Collecting huggingface-hub==0.0.8 Downloading https://files.pythonhosted.org/packages/a1/88/7b1e45720ecf59c6c6737ff332f41c955963090a18e72acbcbeac6b25e86/huggingface_hub-0.0.8-py3-none-any.whl Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.25.1) Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.60.0) Collecting tokenizers<0.11,>=0.10.1 Downloading https://files.pythonhosted.org/packages/ae/04/5b870f26a858552025a62f1649c20d29d2672c02ff3c3fb4c688ca46467a/tokenizers-0.10.2-cp37-cp37m-manylinux2010_x86_64.whl (3.3MB) |████████████████████████████████| 3.3MB 50.3MB/s Requirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.0.12) Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2019.12.20) Requirement already satisfied: packaging in /usr/local/lib/python3.7/dist-packages (from transformers) (20.9) Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata; python_version < "3.8"->transformers) (3.4.1) Requirement already satisfied: typing-extensions>=3.6.4; python_version < "3.8" in /usr/local/lib/python3.7/dist-packages (from importlib-metadata; python_version < "3.8"->transformers) (3.7.4.3) Requirement already satisfied: joblib in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.0.1) Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.15.0) Requirement already satisfied: click in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (7.1.2) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2020.12.5) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10) Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3) Requirement already satisfied: chardet<5,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4) Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging->transformers) (2.4.7) Installing collected packages: sacremoses, huggingface-hub, tokenizers, transformers Successfully installed huggingface-hub-0.0.8 sacremoses-0.0.45 tokenizers-0.10.2 transformers-4.6.0
Run the WILDS package¶
This sample command will train a FasterRCNN model with the Group DRO algorithm that tries to minimize the loss of the worst-case training domain.
To train using a different algorithm like standard empirical risk minimization (ERM), simply replace --algorithm groupDRO with --algorithm ERM.
WILDS will automatically handle downloading the Global Wheat Competition dataset to the location specified in root_dir. For the purposes of this competition, both the val and the test sets in the WILDS library do not have labels provided, so the reported val and test accuracies are just dummy values.
In [ ]:
N_EPOCHS=10
SAVE_TO="gdro"
In [ ]:
!python3 wilds/examples/run_expt.py -d globalwheat --algorithm groupDRO --root_dir data --log_dir $SAVE_TO --download --progress_bar --save_step 1 --split_scheme official --n_epochs $N_EPOCHS
2021-05-17 06:49:23.673279: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
Dataset: gwhd
Algorithm: groupDRO
Root dir: data
Split scheme: official
Dataset kwargs: {}
Download: True
Frac: 1.0
Version: None
Loader kwargs: {'num_workers': 1, 'pin_memory': True}
Train loader: standard
Uniform over groups: True
Distinct groups: True
N groups per batch: 4
Batch size: 4
Eval loader: standard
Model: fasterrcnn
Model kwargs: {'n_classes': 1, 'pretrained': True, 'pretrained_model': True, 'pretrained_backbone': True, 'min_size': 1024, 'max_size': 1024}
Train transform: image_base
Eval transform: image_base
Target resolution: None
Resize scale: None
Max token length: None
Loss function: fasterrcnn_criterion
Loss kwargs: {}
Groupby fields: ['location_date_sensor']
Group dro step size: 0.01
Coral penalty weight: None
Irm lambda: None
Irm penalty anneal iters: None
Algo log metric: None
Val metric: detection_acc_avg
Val metric decreasing: False
N epochs: 10
Optimizer: Adam
Lr: 1e-05
Weight decay: 0.0001
Max grad norm: None
Optimizer kwargs: {}
Scheduler: None
Scheduler kwargs: {}
Scheduler metric split: val
Scheduler metric name: None
Process outputs function: None
Evaluate all splits: True
Eval splits: []
Eval only: False
Eval epoch: None
Device: cuda:0
Seed: 0
Log dir: gdro
Log every: 50
Save step: 1
Save best: True
Save last: True
Save pred: True
No group logging: False
Use wandb: False
Progress bar: True
Resume: False
Downloading dataset to data/gwhd_v0.9...
You can also download the dataset manually at https://wilds.stanford.edu/downloads.
Downloading https://worksheets.codalab.org/rest/bundles/0x8ba9122a41454997afdfb78762d390cf/contents/blob/ to data/gwhd_v0.9/archive.tar.gz
100% 10279141376/10280247296 [08:55<00:00, 18152342.55Byte/s]Extracting data/gwhd_v0.9/archive.tar.gz to data/gwhd_v0.9
10280386560Byte [09:43, 17628119.31Byte/s]
It took 13.81 minutes to download and uncompress the dataset.
Train data...
location_date_sensor = 0: n = 29
location_date_sensor = 1: n = 60
location_date_sensor = 2: n = 176
location_date_sensor = 3: n = 20
location_date_sensor = 4: n = 24
location_date_sensor = 5: n = 448
location_date_sensor = 6: n = 160
location_date_sensor = 7: n = 60
location_date_sensor = 8: n = 32
location_date_sensor = 9: n = 82
location_date_sensor = 10: n = 204
location_date_sensor = 11: n = 30
location_date_sensor = 12: n = 747
location_date_sensor = 13: n = 66
location_date_sensor = 14: n = 401
location_date_sensor = 15: n = 588
location_date_sensor = 16: n = 98
location_date_sensor = 17: n = 432
location_date_sensor = 18: n = 0
location_date_sensor = 19: n = 0
location_date_sensor = 20: n = 0
location_date_sensor = 21: n = 0
location_date_sensor = 22: n = 0
location_date_sensor = 23: n = 0
location_date_sensor = 24: n = 0
location_date_sensor = 25: n = 0
location_date_sensor = 26: n = 0
location_date_sensor = 27: n = 0
location_date_sensor = 28: n = 0
location_date_sensor = 29: n = 0
location_date_sensor = 30: n = 0
location_date_sensor = 31: n = 0
location_date_sensor = 32: n = 0
location_date_sensor = 33: n = 0
location_date_sensor = 34: n = 0
location_date_sensor = 35: n = 0
location_date_sensor = 36: n = 0
location_date_sensor = 37: n = 0
location_date_sensor = 38: n = 0
location_date_sensor = 39: n = 0
location_date_sensor = 40: n = 0
location_date_sensor = 41: n = 0
location_date_sensor = 42: n = 0
location_date_sensor = 43: n = 0
location_date_sensor = 44: n = 0
location_date_sensor = 45: n = 0
location_date_sensor = 46: n = 0
Validation data...
location_date_sensor = 0: n = 0
location_date_sensor = 1: n = 0
location_date_sensor = 2: n = 0
location_date_sensor = 3: n = 0
location_date_sensor = 4: n = 0
location_date_sensor = 5: n = 0
location_date_sensor = 6: n = 0
location_date_sensor = 7: n = 0
location_date_sensor = 8: n = 0
location_date_sensor = 9: n = 0
location_date_sensor = 10: n = 0
location_date_sensor = 11: n = 0
location_date_sensor = 12: n = 0
location_date_sensor = 13: n = 0
location_date_sensor = 14: n = 0
location_date_sensor = 15: n = 0
location_date_sensor = 16: n = 0
location_date_sensor = 17: n = 0
location_date_sensor = 18: n = 12
location_date_sensor = 19: n = 49
location_date_sensor = 20: n = 254
location_date_sensor = 21: n = 216
location_date_sensor = 22: n = 89
location_date_sensor = 23: n = 11
location_date_sensor = 24: n = 4
location_date_sensor = 25: n = 7
location_date_sensor = 26: n = 14
location_date_sensor = 27: n = 17
location_date_sensor = 28: n = 14
location_date_sensor = 29: n = 8
location_date_sensor = 30: n = 43
location_date_sensor = 31: n = 55
location_date_sensor = 32: n = 51
location_date_sensor = 33: n = 50
location_date_sensor = 34: n = 28
location_date_sensor = 35: n = 75
location_date_sensor = 36: n = 56
location_date_sensor = 37: n = 34
location_date_sensor = 38: n = 55
location_date_sensor = 39: n = 13
location_date_sensor = 40: n = 19
location_date_sensor = 41: n = 19
location_date_sensor = 42: n = 57
location_date_sensor = 43: n = 39
location_date_sensor = 44: n = 33
location_date_sensor = 45: n = 39
location_date_sensor = 46: n = 30
Test data...
location_date_sensor = 0: n = 0
location_date_sensor = 1: n = 0
location_date_sensor = 2: n = 0
location_date_sensor = 3: n = 0
location_date_sensor = 4: n = 0
location_date_sensor = 5: n = 0
location_date_sensor = 6: n = 0
location_date_sensor = 7: n = 0
location_date_sensor = 8: n = 0
location_date_sensor = 9: n = 0
location_date_sensor = 10: n = 0
location_date_sensor = 11: n = 0
location_date_sensor = 12: n = 0
location_date_sensor = 13: n = 0
location_date_sensor = 14: n = 0
location_date_sensor = 15: n = 0
location_date_sensor = 16: n = 0
location_date_sensor = 17: n = 0
location_date_sensor = 18: n = 8
location_date_sensor = 19: n = 71
location_date_sensor = 20: n = 284
location_date_sensor = 21: n = 240
location_date_sensor = 22: n = 111
location_date_sensor = 23: n = 11
location_date_sensor = 24: n = 12
location_date_sensor = 25: n = 7
location_date_sensor = 26: n = 16
location_date_sensor = 27: n = 13
location_date_sensor = 28: n = 16
location_date_sensor = 29: n = 14
location_date_sensor = 30: n = 57
location_date_sensor = 31: n = 45
location_date_sensor = 32: n = 49
location_date_sensor = 33: n = 45
location_date_sensor = 34: n = 32
location_date_sensor = 35: n = 69
location_date_sensor = 36: n = 49
location_date_sensor = 37: n = 26
location_date_sensor = 38: n = 45
location_date_sensor = 39: n = 4
location_date_sensor = 40: n = 22
location_date_sensor = 41: n = 14
location_date_sensor = 42: n = 49
location_date_sensor = 43: n = 45
location_date_sensor = 44: n = 36
location_date_sensor = 45: n = 38
location_date_sensor = 46: n = 30
Downloading: "https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth" to /root/.cache/torch/hub/checkpoints/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth
100% 160M/160M [00:00<00:00, 193MB/s]
Epoch [0]:
Train:
5% 49/915 [01:11<21:08, 1.46s/it]objective: 0.363
loss_avg: 1.779
location_date_sensor = 0 [n = 14]: weight: 0.059 loss: 1.998
location_date_sensor = 1 [n = 6]: weight: 0.054 loss: 1.548
location_date_sensor = 2 [n = 10]: weight: 0.053 loss: 1.479
location_date_sensor = 3 [n = 5]: weight: 0.056 loss: 2.450
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 1.531
location_date_sensor = 5 [n = 7]: weight: 0.054 loss: 1.603
location_date_sensor = 6 [n = 9]: weight: 0.055 loss: 1.743
location_date_sensor = 7 [n = 14]: weight: 0.058 loss: 2.052
location_date_sensor = 8 [n = 12]: weight: 0.060 loss: 2.311
location_date_sensor = 9 [n = 14]: weight: 0.063 loss: 2.453
location_date_sensor = 10 [n = 12]: weight: 0.054 loss: 1.401
location_date_sensor = 11 [n = 11]: weight: 0.059 loss: 2.345
location_date_sensor = 12 [n = 10]: weight: 0.053 loss: 1.396
location_date_sensor = 13 [n = 12]: weight: 0.055 loss: 1.569
location_date_sensor = 14 [n = 11]: weight: 0.057 loss: 1.775
location_date_sensor = 15 [n = 14]: weight: 0.054 loss: 1.516
location_date_sensor = 16 [n = 19]: weight: 0.056 loss: 1.451
location_date_sensor = 17 [n = 9]: weight: 0.055 loss: 1.474
11% 99/915 [02:27<20:39, 1.52s/it]objective: 0.263
loss_avg: 1.289
location_date_sensor = 0 [n = 8]: weight: 0.058 loss: 1.654
location_date_sensor = 1 [n = 6]: weight: 0.048 loss: 0.929
location_date_sensor = 2 [n = 15]: weight: 0.053 loss: 1.236
location_date_sensor = 3 [n = 16]: weight: 0.055 loss: 1.333
location_date_sensor = 4 [n = 10]: weight: 0.054 loss: 1.280
location_date_sensor = 5 [n = 10]: weight: 0.051 loss: 0.900
location_date_sensor = 6 [n = 19]: weight: 0.057 loss: 1.177
location_date_sensor = 7 [n = 15]: weight: 0.065 loss: 1.693
location_date_sensor = 8 [n = 10]: weight: 0.058 loss: 1.351
location_date_sensor = 9 [n = 8]: weight: 0.063 loss: 1.397
location_date_sensor = 10 [n = 8]: weight: 0.052 loss: 1.217
location_date_sensor = 11 [n = 7]: weight: 0.058 loss: 1.600
location_date_sensor = 12 [n = 11]: weight: 0.054 loss: 1.211
location_date_sensor = 13 [n = 10]: weight: 0.055 loss: 1.325
location_date_sensor = 14 [n = 11]: weight: 0.056 loss: 1.303
location_date_sensor = 15 [n = 18]: weight: 0.058 loss: 1.326
location_date_sensor = 16 [n = 10]: weight: 0.058 loss: 1.131
location_date_sensor = 17 [n = 8]: weight: 0.050 loss: 0.989
16% 149/915 [03:43<19:30, 1.53s/it]objective: 0.237
loss_avg: 1.136
location_date_sensor = 0 [n = 14]: weight: 0.060 loss: 1.224
location_date_sensor = 1 [n = 13]: weight: 0.047 loss: 0.782
location_date_sensor = 2 [n = 11]: weight: 0.054 loss: 1.129
location_date_sensor = 3 [n = 10]: weight: 0.055 loss: 1.096
location_date_sensor = 4 [n = 14]: weight: 0.054 loss: 1.233
location_date_sensor = 5 [n = 12]: weight: 0.049 loss: 0.940
location_date_sensor = 6 [n = 14]: weight: 0.057 loss: 1.005
location_date_sensor = 7 [n = 14]: weight: 0.070 loss: 1.451
location_date_sensor = 8 [n = 12]: weight: 0.059 loss: 1.205
location_date_sensor = 9 [n = 12]: weight: 0.062 loss: 1.305
location_date_sensor = 10 [n = 13]: weight: 0.052 loss: 1.186
location_date_sensor = 11 [n = 9]: weight: 0.059 loss: 1.461
location_date_sensor = 12 [n = 9]: weight: 0.052 loss: 1.014
location_date_sensor = 13 [n = 10]: weight: 0.056 loss: 1.107
location_date_sensor = 14 [n = 8]: weight: 0.056 loss: 1.147
location_date_sensor = 15 [n = 8]: weight: 0.060 loss: 1.214
location_date_sensor = 16 [n = 5]: weight: 0.053 loss: 1.095
location_date_sensor = 17 [n = 12]: weight: 0.049 loss: 0.864
22% 199/915 [05:00<18:10, 1.52s/it]objective: 0.225
loss_avg: 1.067
location_date_sensor = 0 [n = 10]: weight: 0.062 loss: 1.331
location_date_sensor = 1 [n = 10]: weight: 0.045 loss: 0.749
location_date_sensor = 2 [n = 13]: weight: 0.054 loss: 0.953
location_date_sensor = 3 [n = 9]: weight: 0.055 loss: 1.025
location_date_sensor = 4 [n = 10]: weight: 0.055 loss: 1.034
location_date_sensor = 5 [n = 11]: weight: 0.048 loss: 0.922
location_date_sensor = 6 [n = 8]: weight: 0.057 loss: 0.927
location_date_sensor = 7 [n = 10]: weight: 0.070 loss: 1.337
location_date_sensor = 8 [n = 12]: weight: 0.060 loss: 1.063
location_date_sensor = 9 [n = 19]: weight: 0.065 loss: 1.125
location_date_sensor = 10 [n = 18]: weight: 0.055 loss: 1.108
location_date_sensor = 11 [n = 8]: weight: 0.058 loss: 1.340
location_date_sensor = 12 [n = 10]: weight: 0.050 loss: 0.961
location_date_sensor = 13 [n = 10]: weight: 0.056 loss: 1.198
location_date_sensor = 14 [n = 12]: weight: 0.055 loss: 1.082
location_date_sensor = 15 [n = 16]: weight: 0.059 loss: 1.057
location_date_sensor = 16 [n = 8]: weight: 0.053 loss: 1.093
location_date_sensor = 17 [n = 6]: weight: 0.047 loss: 0.767
27% 249/915 [06:16<16:53, 1.52s/it]objective: 0.220
loss_avg: 1.020
location_date_sensor = 0 [n = 9]: weight: 0.064 loss: 1.383
location_date_sensor = 1 [n = 15]: weight: 0.043 loss: 0.725
location_date_sensor = 2 [n = 10]: weight: 0.054 loss: 0.946
location_date_sensor = 3 [n = 7]: weight: 0.053 loss: 1.022
location_date_sensor = 4 [n = 8]: weight: 0.053 loss: 1.043
location_date_sensor = 5 [n = 7]: weight: 0.046 loss: 0.776
location_date_sensor = 6 [n = 14]: weight: 0.056 loss: 0.896
location_date_sensor = 7 [n = 22]: weight: 0.077 loss: 1.249
location_date_sensor = 8 [n = 6]: weight: 0.060 loss: 1.019
location_date_sensor = 9 [n = 15]: weight: 0.068 loss: 1.094
location_date_sensor = 10 [n = 15]: weight: 0.057 loss: 1.041
location_date_sensor = 11 [n = 11]: weight: 0.057 loss: 1.247
location_date_sensor = 12 [n = 8]: weight: 0.049 loss: 0.915
location_date_sensor = 13 [n = 11]: weight: 0.054 loss: 1.056
location_date_sensor = 14 [n = 8]: weight: 0.055 loss: 1.056
location_date_sensor = 15 [n = 10]: weight: 0.060 loss: 0.996
location_date_sensor = 16 [n = 12]: weight: 0.052 loss: 1.027
location_date_sensor = 17 [n = 12]: weight: 0.044 loss: 0.737
33% 299/915 [07:32<15:40, 1.53s/it]objective: 0.195
loss_avg: 0.954
location_date_sensor = 0 [n = 7]: weight: 0.063 loss: 0.891
location_date_sensor = 1 [n = 11]: weight: 0.042 loss: 0.624
location_date_sensor = 2 [n = 13]: weight: 0.053 loss: 0.940
location_date_sensor = 3 [n = 10]: weight: 0.052 loss: 0.962
location_date_sensor = 4 [n = 8]: weight: 0.053 loss: 0.953
location_date_sensor = 5 [n = 10]: weight: 0.045 loss: 0.859
location_date_sensor = 6 [n = 15]: weight: 0.057 loss: 0.908
location_date_sensor = 7 [n = 12]: weight: 0.080 loss: 1.122
location_date_sensor = 8 [n = 15]: weight: 0.059 loss: 0.987
location_date_sensor = 9 [n = 17]: weight: 0.071 loss: 1.023
location_date_sensor = 10 [n = 8]: weight: 0.058 loss: 1.118
location_date_sensor = 11 [n = 10]: weight: 0.058 loss: 1.186
location_date_sensor = 12 [n = 5]: weight: 0.046 loss: 0.890
location_date_sensor = 13 [n = 8]: weight: 0.053 loss: 1.130
location_date_sensor = 14 [n = 15]: weight: 0.056 loss: 0.998
location_date_sensor = 15 [n = 8]: weight: 0.061 loss: 1.001
location_date_sensor = 16 [n = 13]: weight: 0.053 loss: 0.953
location_date_sensor = 17 [n = 15]: weight: 0.044 loss: 0.713
38% 349/915 [08:49<14:24, 1.53s/it]objective: 0.201
loss_avg: 0.960
location_date_sensor = 0 [n = 14]: weight: 0.063 loss: 1.040
location_date_sensor = 1 [n = 12]: weight: 0.041 loss: 0.695
location_date_sensor = 2 [n = 13]: weight: 0.053 loss: 0.795
location_date_sensor = 3 [n = 10]: weight: 0.050 loss: 0.870
location_date_sensor = 4 [n = 9]: weight: 0.051 loss: 0.888
location_date_sensor = 5 [n = 13]: weight: 0.045 loss: 0.728
location_date_sensor = 6 [n = 9]: weight: 0.057 loss: 0.889
location_date_sensor = 7 [n = 14]: weight: 0.085 loss: 1.135
location_date_sensor = 8 [n = 4]: weight: 0.057 loss: 1.032
location_date_sensor = 9 [n = 12]: weight: 0.074 loss: 1.271
location_date_sensor = 10 [n = 6]: weight: 0.056 loss: 1.024
location_date_sensor = 11 [n = 6]: weight: 0.057 loss: 1.185
location_date_sensor = 12 [n = 16]: weight: 0.046 loss: 0.858
location_date_sensor = 13 [n = 12]: weight: 0.054 loss: 1.057
location_date_sensor = 14 [n = 12]: weight: 0.056 loss: 1.017
location_date_sensor = 15 [n = 14]: weight: 0.060 loss: 1.019
location_date_sensor = 16 [n = 15]: weight: 0.054 loss: 1.137
location_date_sensor = 17 [n = 9]: weight: 0.043 loss: 0.680
44% 399/915 [10:05<13:06, 1.52s/it]objective: 0.196
loss_avg: 0.933
location_date_sensor = 0 [n = 11]: weight: 0.064 loss: 0.944
location_date_sensor = 1 [n = 10]: weight: 0.040 loss: 0.650
location_date_sensor = 2 [n = 11]: weight: 0.054 loss: 0.851
location_date_sensor = 3 [n = 10]: weight: 0.050 loss: 0.900
location_date_sensor = 4 [n = 15]: weight: 0.051 loss: 0.985
location_date_sensor = 5 [n = 18]: weight: 0.045 loss: 0.849
location_date_sensor = 6 [n = 10]: weight: 0.057 loss: 0.862
location_date_sensor = 7 [n = 8]: weight: 0.083 loss: 1.043
location_date_sensor = 8 [n = 5]: weight: 0.056 loss: 0.934
location_date_sensor = 9 [n = 12]: weight: 0.077 loss: 1.046
location_date_sensor = 10 [n = 10]: weight: 0.054 loss: 0.994
location_date_sensor = 11 [n = 12]: weight: 0.056 loss: 1.110
location_date_sensor = 12 [n = 7]: weight: 0.046 loss: 0.889
location_date_sensor = 13 [n = 8]: weight: 0.054 loss: 1.070
location_date_sensor = 14 [n = 10]: weight: 0.057 loss: 1.075
location_date_sensor = 15 [n = 17]: weight: 0.063 loss: 0.998
location_date_sensor = 16 [n = 11]: weight: 0.055 loss: 0.958
location_date_sensor = 17 [n = 15]: weight: 0.042 loss: 0.724
49% 449/915 [11:21<11:51, 1.53s/it]objective: 0.193
loss_avg: 0.914
location_date_sensor = 0 [n = 9]: weight: 0.063 loss: 0.864
location_date_sensor = 1 [n = 11]: weight: 0.038 loss: 0.622
location_date_sensor = 2 [n = 7]: weight: 0.052 loss: 0.822
location_date_sensor = 3 [n = 7]: weight: 0.049 loss: 0.846
location_date_sensor = 4 [n = 15]: weight: 0.053 loss: 0.969
location_date_sensor = 5 [n = 12]: weight: 0.046 loss: 0.614
location_date_sensor = 6 [n = 8]: weight: 0.055 loss: 0.775
location_date_sensor = 7 [n = 13]: weight: 0.087 loss: 1.213
location_date_sensor = 8 [n = 12]: weight: 0.054 loss: 0.920
location_date_sensor = 9 [n = 17]: weight: 0.080 loss: 1.058
location_date_sensor = 10 [n = 13]: weight: 0.055 loss: 1.034
location_date_sensor = 11 [n = 9]: weight: 0.057 loss: 1.149
location_date_sensor = 12 [n = 8]: weight: 0.045 loss: 0.825
location_date_sensor = 13 [n = 8]: weight: 0.053 loss: 1.016
location_date_sensor = 14 [n = 15]: weight: 0.059 loss: 0.980
location_date_sensor = 15 [n = 9]: weight: 0.064 loss: 0.882
location_date_sensor = 16 [n = 15]: weight: 0.055 loss: 0.950
location_date_sensor = 17 [n = 12]: weight: 0.041 loss: 0.680
55% 499/915 [12:38<10:35, 1.53s/it]objective: 0.200
loss_avg: 0.915
location_date_sensor = 0 [n = 11]: weight: 0.062 loss: 0.897
location_date_sensor = 1 [n = 9]: weight: 0.037 loss: 0.590
location_date_sensor = 2 [n = 12]: weight: 0.051 loss: 0.811
location_date_sensor = 3 [n = 9]: weight: 0.048 loss: 0.887
location_date_sensor = 4 [n = 10]: weight: 0.054 loss: 0.956
location_date_sensor = 5 [n = 18]: weight: 0.045 loss: 0.818
location_date_sensor = 6 [n = 13]: weight: 0.053 loss: 0.828
location_date_sensor = 7 [n = 15]: weight: 0.091 loss: 1.093
location_date_sensor = 8 [n = 10]: weight: 0.053 loss: 0.954
location_date_sensor = 9 [n = 11]: weight: 0.083 loss: 1.046
location_date_sensor = 10 [n = 11]: weight: 0.056 loss: 1.019
location_date_sensor = 11 [n = 11]: weight: 0.057 loss: 1.099
location_date_sensor = 12 [n = 5]: weight: 0.041 loss: 0.870
location_date_sensor = 13 [n = 15]: weight: 0.054 loss: 1.053
location_date_sensor = 14 [n = 8]: weight: 0.059 loss: 0.982
location_date_sensor = 15 [n = 12]: weight: 0.064 loss: 0.903
location_date_sensor = 16 [n = 8]: weight: 0.055 loss: 0.982
location_date_sensor = 17 [n = 12]: weight: 0.040 loss: 0.641
60% 549/915 [13:54<09:17, 1.52s/it]objective: 0.189
loss_avg: 0.883
location_date_sensor = 0 [n = 11]: weight: 0.061 loss: 0.995
location_date_sensor = 1 [n = 7]: weight: 0.035 loss: 0.625
location_date_sensor = 2 [n = 10]: weight: 0.050 loss: 0.794
location_date_sensor = 3 [n = 12]: weight: 0.047 loss: 0.762
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 0.940
location_date_sensor = 5 [n = 12]: weight: 0.046 loss: 0.744
location_date_sensor = 6 [n = 12]: weight: 0.054 loss: 0.860
location_date_sensor = 7 [n = 12]: weight: 0.095 loss: 1.077
location_date_sensor = 8 [n = 10]: weight: 0.053 loss: 0.893
location_date_sensor = 9 [n = 15]: weight: 0.086 loss: 0.917
location_date_sensor = 10 [n = 11]: weight: 0.056 loss: 1.016
location_date_sensor = 11 [n = 8]: weight: 0.057 loss: 0.994
location_date_sensor = 12 [n = 14]: weight: 0.041 loss: 0.817
location_date_sensor = 13 [n = 11]: weight: 0.055 loss: 1.052
location_date_sensor = 14 [n = 5]: weight: 0.058 loss: 1.021
location_date_sensor = 15 [n = 13]: weight: 0.064 loss: 0.906
location_date_sensor = 16 [n = 13]: weight: 0.055 loss: 0.848
location_date_sensor = 17 [n = 13]: weight: 0.039 loss: 0.683
65% 599/915 [15:10<08:03, 1.53s/it]objective: 0.200
loss_avg: 0.904
location_date_sensor = 0 [n = 8]: weight: 0.061 loss: 0.866
location_date_sensor = 1 [n = 4]: weight: 0.033 loss: 0.682
location_date_sensor = 2 [n = 10]: weight: 0.049 loss: 0.863
location_date_sensor = 3 [n = 13]: weight: 0.047 loss: 0.752
location_date_sensor = 4 [n = 15]: weight: 0.055 loss: 0.914
location_date_sensor = 5 [n = 9]: weight: 0.045 loss: 0.829
location_date_sensor = 6 [n = 10]: weight: 0.052 loss: 0.729
location_date_sensor = 7 [n = 12]: weight: 0.097 loss: 1.025
location_date_sensor = 8 [n = 8]: weight: 0.051 loss: 0.914
location_date_sensor = 9 [n = 14]: weight: 0.088 loss: 0.974
location_date_sensor = 10 [n = 15]: weight: 0.058 loss: 1.072
location_date_sensor = 11 [n = 12]: weight: 0.056 loss: 1.044
location_date_sensor = 12 [n = 10]: weight: 0.041 loss: 0.824
location_date_sensor = 13 [n = 18]: weight: 0.059 loss: 0.951
location_date_sensor = 14 [n = 11]: weight: 0.056 loss: 0.989
location_date_sensor = 15 [n = 12]: weight: 0.065 loss: 0.952
location_date_sensor = 16 [n = 7]: weight: 0.054 loss: 0.909
location_date_sensor = 17 [n = 12]: weight: 0.038 loss: 0.698
71% 649/915 [16:26<06:45, 1.52s/it]objective: 0.182
loss_avg: 0.885
location_date_sensor = 0 [n = 8]: weight: 0.060 loss: 0.785
location_date_sensor = 1 [n = 11]: weight: 0.031 loss: 0.628
location_date_sensor = 2 [n = 10]: weight: 0.048 loss: 0.743
location_date_sensor = 3 [n = 4]: weight: 0.044 loss: 0.873
location_date_sensor = 4 [n = 5]: weight: 0.054 loss: 0.844
location_date_sensor = 5 [n = 7]: weight: 0.043 loss: 0.752
location_date_sensor = 6 [n = 13]: weight: 0.052 loss: 0.825
location_date_sensor = 7 [n = 7]: weight: 0.096 loss: 1.145
location_date_sensor = 8 [n = 11]: weight: 0.051 loss: 0.945
location_date_sensor = 9 [n = 7]: weight: 0.089 loss: 0.951
location_date_sensor = 10 [n = 13]: weight: 0.059 loss: 0.978
location_date_sensor = 11 [n = 15]: weight: 0.058 loss: 1.037
location_date_sensor = 12 [n = 12]: weight: 0.040 loss: 0.808
location_date_sensor = 13 [n = 10]: weight: 0.061 loss: 1.016
location_date_sensor = 14 [n = 20]: weight: 0.059 loss: 0.948
location_date_sensor = 15 [n = 19]: weight: 0.067 loss: 0.956
location_date_sensor = 16 [n = 13]: weight: 0.054 loss: 1.001
location_date_sensor = 17 [n = 15]: weight: 0.037 loss: 0.621
76% 699/915 [17:43<05:30, 1.53s/it]objective: 0.184
loss_avg: 0.876
location_date_sensor = 0 [n = 13]: weight: 0.059 loss: 0.808
location_date_sensor = 1 [n = 6]: weight: 0.030 loss: 0.648
location_date_sensor = 2 [n = 12]: weight: 0.048 loss: 0.782
location_date_sensor = 3 [n = 11]: weight: 0.043 loss: 0.828
location_date_sensor = 4 [n = 9]: weight: 0.053 loss: 0.884
location_date_sensor = 5 [n = 13]: weight: 0.043 loss: 0.768
location_date_sensor = 6 [n = 11]: weight: 0.051 loss: 0.777
location_date_sensor = 7 [n = 9]: weight: 0.096 loss: 1.061
location_date_sensor = 8 [n = 13]: weight: 0.051 loss: 0.865
location_date_sensor = 9 [n = 8]: weight: 0.086 loss: 0.931
location_date_sensor = 10 [n = 14]: weight: 0.060 loss: 1.002
location_date_sensor = 11 [n = 10]: weight: 0.060 loss: 0.981
location_date_sensor = 12 [n = 13]: weight: 0.039 loss: 0.815
location_date_sensor = 13 [n = 16]: weight: 0.062 loss: 0.997
location_date_sensor = 14 [n = 15]: weight: 0.062 loss: 0.967
location_date_sensor = 15 [n = 8]: weight: 0.068 loss: 0.928
location_date_sensor = 16 [n = 10]: weight: 0.055 loss: 0.933
location_date_sensor = 17 [n = 9]: weight: 0.036 loss: 0.661
82% 749/915 [18:59<04:13, 1.53s/it]objective: 0.180
loss_avg: 0.834
location_date_sensor = 0 [n = 15]: weight: 0.059 loss: 0.785
location_date_sensor = 1 [n = 10]: weight: 0.029 loss: 0.581
location_date_sensor = 2 [n = 17]: weight: 0.048 loss: 0.729
location_date_sensor = 3 [n = 11]: weight: 0.043 loss: 0.770
location_date_sensor = 4 [n = 11]: weight: 0.052 loss: 0.920
location_date_sensor = 5 [n = 7]: weight: 0.041 loss: 0.869
location_date_sensor = 6 [n = 9]: weight: 0.051 loss: 0.768
location_date_sensor = 7 [n = 12]: weight: 0.097 loss: 0.973
location_date_sensor = 8 [n = 9]: weight: 0.051 loss: 0.856
location_date_sensor = 9 [n = 14]: weight: 0.088 loss: 0.953
location_date_sensor = 10 [n = 7]: weight: 0.060 loss: 1.001
location_date_sensor = 11 [n = 6]: weight: 0.059 loss: 1.069
location_date_sensor = 12 [n = 12]: weight: 0.039 loss: 0.794
location_date_sensor = 13 [n = 14]: weight: 0.066 loss: 0.929
location_date_sensor = 14 [n = 11]: weight: 0.065 loss: 0.950
location_date_sensor = 15 [n = 11]: weight: 0.067 loss: 0.874
location_date_sensor = 16 [n = 9]: weight: 0.054 loss: 0.734
location_date_sensor = 17 [n = 15]: weight: 0.036 loss: 0.660
87% 799/915 [20:15<02:57, 1.53s/it]objective: 0.179
loss_avg: 0.858
location_date_sensor = 0 [n = 12]: weight: 0.059 loss: 0.994
location_date_sensor = 1 [n = 12]: weight: 0.028 loss: 0.590
location_date_sensor = 2 [n = 8]: weight: 0.047 loss: 0.806
location_date_sensor = 3 [n = 15]: weight: 0.044 loss: 0.838
location_date_sensor = 4 [n = 15]: weight: 0.053 loss: 0.864
location_date_sensor = 5 [n = 10]: weight: 0.040 loss: 0.783
location_date_sensor = 6 [n = 8]: weight: 0.049 loss: 0.766
location_date_sensor = 7 [n = 7]: weight: 0.096 loss: 0.838
location_date_sensor = 8 [n = 11]: weight: 0.052 loss: 0.815
location_date_sensor = 9 [n = 13]: weight: 0.092 loss: 0.924
location_date_sensor = 10 [n = 11]: weight: 0.060 loss: 0.924
location_date_sensor = 11 [n = 16]: weight: 0.058 loss: 0.982
location_date_sensor = 12 [n = 13]: weight: 0.039 loss: 0.804
location_date_sensor = 13 [n = 9]: weight: 0.066 loss: 1.053
location_date_sensor = 14 [n = 11]: weight: 0.065 loss: 0.912
location_date_sensor = 15 [n = 12]: weight: 0.068 loss: 0.938
location_date_sensor = 16 [n = 6]: weight: 0.052 loss: 0.950
location_date_sensor = 17 [n = 11]: weight: 0.036 loss: 0.641
93% 849/915 [21:32<01:40, 1.52s/it]objective: 0.172
loss_avg: 0.843
location_date_sensor = 0 [n = 5]: weight: 0.059 loss: 0.865
location_date_sensor = 1 [n = 13]: weight: 0.028 loss: 0.568
location_date_sensor = 2 [n = 11]: weight: 0.046 loss: 0.681
location_date_sensor = 3 [n = 14]: weight: 0.046 loss: 0.781
location_date_sensor = 4 [n = 11]: weight: 0.054 loss: 0.865
location_date_sensor = 5 [n = 15]: weight: 0.041 loss: 0.797
location_date_sensor = 6 [n = 13]: weight: 0.048 loss: 0.823
location_date_sensor = 7 [n = 9]: weight: 0.093 loss: 0.918
location_date_sensor = 8 [n = 10]: weight: 0.050 loss: 0.875
location_date_sensor = 9 [n = 9]: weight: 0.091 loss: 0.903
location_date_sensor = 10 [n = 7]: weight: 0.060 loss: 0.921
location_date_sensor = 11 [n = 9]: weight: 0.059 loss: 0.956
location_date_sensor = 12 [n = 13]: weight: 0.039 loss: 0.807
location_date_sensor = 13 [n = 19]: weight: 0.069 loss: 0.963
location_date_sensor = 14 [n = 14]: weight: 0.067 loss: 0.951
location_date_sensor = 15 [n = 7]: weight: 0.067 loss: 0.790
location_date_sensor = 16 [n = 15]: weight: 0.052 loss: 0.979
location_date_sensor = 17 [n = 6]: weight: 0.034 loss: 0.647
98% 899/915 [22:48<00:24, 1.52s/it]objective: 0.180
loss_avg: 0.834
location_date_sensor = 0 [n = 14]: weight: 0.058 loss: 0.949
location_date_sensor = 1 [n = 10]: weight: 0.027 loss: 0.588
location_date_sensor = 2 [n = 5]: weight: 0.045 loss: 0.713
location_date_sensor = 3 [n = 14]: weight: 0.046 loss: 0.719
location_date_sensor = 4 [n = 8]: weight: 0.053 loss: 0.809
location_date_sensor = 5 [n = 10]: weight: 0.041 loss: 0.683
location_date_sensor = 6 [n = 10]: weight: 0.048 loss: 0.796
location_date_sensor = 7 [n = 14]: weight: 0.095 loss: 0.940
location_date_sensor = 8 [n = 14]: weight: 0.052 loss: 0.816
location_date_sensor = 9 [n = 16]: weight: 0.093 loss: 0.876
location_date_sensor = 10 [n = 11]: weight: 0.059 loss: 0.974
location_date_sensor = 11 [n = 10]: weight: 0.060 loss: 0.950
location_date_sensor = 12 [n = 16]: weight: 0.040 loss: 0.792
location_date_sensor = 13 [n = 13]: weight: 0.072 loss: 0.906
location_date_sensor = 14 [n = 9]: weight: 0.066 loss: 0.935
location_date_sensor = 15 [n = 6]: weight: 0.063 loss: 0.931
location_date_sensor = 16 [n = 11]: weight: 0.053 loss: 0.863
location_date_sensor = 17 [n = 9]: weight: 0.033 loss: 0.671
100% 915/915 [23:12<00:00, 1.52s/it]
objective: 0.171
loss_avg: 0.845
location_date_sensor = 0 [n = 2]: weight: 0.057 loss: 0.668
location_date_sensor = 1 [n = 1]: weight: 0.026 loss: 0.738
location_date_sensor = 2 [n = 5]: weight: 0.044 loss: 0.746
location_date_sensor = 3 [n = 7]: weight: 0.047 loss: 0.770
location_date_sensor = 4 [n = 3]: weight: 0.053 loss: 0.835
location_date_sensor = 5 [n = 3]: weight: 0.040 loss: 0.900
location_date_sensor = 6 [n = 1]: weight: 0.046 loss: 0.799
location_date_sensor = 7 [n = 7]: weight: 0.098 loss: 0.921
location_date_sensor = 8 [n = 2]: weight: 0.052 loss: 0.835
location_date_sensor = 9 [n = 2]: weight: 0.095 loss: 1.050
location_date_sensor = 10 [n = 1]: weight: 0.059 loss: 0.912
location_date_sensor = 11 [n = 5]: weight: 0.060 loss: 0.983
location_date_sensor = 12 [n = 4]: weight: 0.041 loss: 0.735
location_date_sensor = 13 [n = 4]: weight: 0.072 loss: 0.912
location_date_sensor = 14 [n = 2]: weight: 0.066 loss: 0.835
location_date_sensor = 15 [n = 3]: weight: 0.064 loss: 0.800
location_date_sensor = 16 [n = 3]: weight: 0.052 loss: 0.996
location_date_sensor = 17 [n = 2]: weight: 0.032 loss: 0.625
Epoch eval:
Average detection_acc: 0.617
location_date_sensor = 0 [n = 195]: detection_acc = 0.601
location_date_sensor = 1 [n = 177]: detection_acc = 0.799
location_date_sensor = 2 [n = 203]: detection_acc = 0.736
location_date_sensor = 3 [n = 194]: detection_acc = 0.666
location_date_sensor = 4 [n = 198]: detection_acc = 0.622
location_date_sensor = 5 [n = 204]: detection_acc = 0.666
location_date_sensor = 6 [n = 206]: detection_acc = 0.668
location_date_sensor = 7 [n = 226]: detection_acc = 0.474
location_date_sensor = 8 [n = 186]: detection_acc = 0.590
location_date_sensor = 9 [n = 235]: detection_acc = 0.481
location_date_sensor = 10 [n = 204]: detection_acc = 0.573
location_date_sensor = 11 [n = 185]: detection_acc = 0.457
location_date_sensor = 12 [n = 196]: detection_acc = 0.645
location_date_sensor = 13 [n = 218]: detection_acc = 0.598
location_date_sensor = 14 [n = 208]: detection_acc = 0.643
location_date_sensor = 15 [n = 217]: detection_acc = 0.648
location_date_sensor = 16 [n = 203]: detection_acc = 0.529
location_date_sensor = 17 [n = 202]: detection_acc = 0.763
Worst-group detection_acc: 0.457
Validation:
100% 348/348 [03:19<00:00, 1.75it/s]
objective: 0.000
loss_avg: 1.000
location_date_sensor = 18 [n = 12]: weight: 0.000 loss: 1.000
location_date_sensor = 19 [n = 49]: weight: 0.000 loss: 1.000
location_date_sensor = 20 [n = 254]: weight: 0.000 loss: 1.000
location_date_sensor = 21 [n = 216]: weight: 0.000 loss: 1.000
location_date_sensor = 22 [n = 89]: weight: 0.000 loss: 1.000
location_date_sensor = 23 [n = 11]: weight: 0.000 loss: 1.000
location_date_sensor = 24 [n = 4]: weight: 0.000 loss: 1.000
location_date_sensor = 25 [n = 7]: weight: 0.000 loss: 1.000
location_date_sensor = 26 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 27 [n = 17]: weight: 0.000 loss: 1.000
location_date_sensor = 28 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 29 [n = 8]: weight: 0.000 loss: 1.000
location_date_sensor = 30 [n = 43]: weight: 0.000 loss: 1.000
location_date_sensor = 31 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 32 [n = 51]: weight: 0.000 loss: 1.000
location_date_sensor = 33 [n = 50]: weight: 0.000 loss: 1.000
location_date_sensor = 34 [n = 28]: weight: 0.000 loss: 1.000
location_date_sensor = 35 [n = 75]: weight: 0.000 loss: 1.000
location_date_sensor = 36 [n = 56]: weight: 0.000 loss: 1.000
location_date_sensor = 37 [n = 34]: weight: 0.000 loss: 1.000
location_date_sensor = 38 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 39 [n = 13]: weight: 0.000 loss: 1.000
location_date_sensor = 40 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 41 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 42 [n = 57]: weight: 0.000 loss: 1.000
location_date_sensor = 43 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 44 [n = 33]: weight: 0.000 loss: 1.000
location_date_sensor = 45 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 46 [n = 30]: weight: 0.000 loss: 1.000
Epoch eval:
Average detection_acc: 0.007
location_date_sensor = 18 [n = 12]: detection_acc = 0.000
location_date_sensor = 19 [n = 49]: detection_acc = 0.020
location_date_sensor = 20 [n = 254]: detection_acc = 0.000
location_date_sensor = 21 [n = 216]: detection_acc = 0.000
location_date_sensor = 22 [n = 89]: detection_acc = 0.000
location_date_sensor = 23 [n = 11]: detection_acc = 0.000
location_date_sensor = 24 [n = 4]: detection_acc = 0.750
location_date_sensor = 25 [n = 7]: detection_acc = 0.000
location_date_sensor = 26 [n = 14]: detection_acc = 0.000
location_date_sensor = 27 [n = 17]: detection_acc = 0.059
location_date_sensor = 28 [n = 14]: detection_acc = 0.000
location_date_sensor = 29 [n = 8]: detection_acc = 0.000
location_date_sensor = 30 [n = 43]: detection_acc = 0.000
location_date_sensor = 31 [n = 55]: detection_acc = 0.000
location_date_sensor = 32 [n = 51]: detection_acc = 0.000
location_date_sensor = 33 [n = 50]: detection_acc = 0.000
location_date_sensor = 34 [n = 28]: detection_acc = 0.000
location_date_sensor = 35 [n = 75]: detection_acc = 0.013
location_date_sensor = 36 [n = 56]: detection_acc = 0.018
location_date_sensor = 37 [n = 34]: detection_acc = 0.029
location_date_sensor = 38 [n = 55]: detection_acc = 0.000
location_date_sensor = 39 [n = 13]: detection_acc = 0.000
location_date_sensor = 40 [n = 19]: detection_acc = 0.000
location_date_sensor = 41 [n = 19]: detection_acc = 0.000
location_date_sensor = 42 [n = 57]: detection_acc = 0.000
location_date_sensor = 43 [n = 39]: detection_acc = 0.000
location_date_sensor = 44 [n = 33]: detection_acc = 0.030
location_date_sensor = 45 [n = 39]: detection_acc = 0.000
location_date_sensor = 46 [n = 30]: detection_acc = 0.033
Worst-group detection_acc: 0.000
Validation detection_acc_avg: 0.007
Epoch 0 has the best validation performance so far.
100% 365/365 [03:28<00:00, 1.75it/s]
Epoch [1]:
Train:
5% 49/915 [01:15<22:09, 1.54s/it]objective: 0.174
loss_avg: 0.814
location_date_sensor = 0 [n = 10]: weight: 0.057 loss: 0.795
location_date_sensor = 1 [n = 11]: weight: 0.025 loss: 0.587
location_date_sensor = 2 [n = 8]: weight: 0.043 loss: 0.720
location_date_sensor = 3 [n = 15]: weight: 0.047 loss: 0.730
location_date_sensor = 4 [n = 14]: weight: 0.053 loss: 0.828
location_date_sensor = 5 [n = 12]: weight: 0.040 loss: 0.766
location_date_sensor = 6 [n = 8]: weight: 0.046 loss: 0.730
location_date_sensor = 7 [n = 9]: weight: 0.099 loss: 0.997
location_date_sensor = 8 [n = 10]: weight: 0.050 loss: 0.795
location_date_sensor = 9 [n = 17]: weight: 0.096 loss: 0.900
location_date_sensor = 10 [n = 8]: weight: 0.059 loss: 0.978
location_date_sensor = 11 [n = 15]: weight: 0.061 loss: 0.956
location_date_sensor = 12 [n = 12]: weight: 0.041 loss: 0.714
location_date_sensor = 13 [n = 13]: weight: 0.074 loss: 0.947
location_date_sensor = 14 [n = 10]: weight: 0.067 loss: 0.969
location_date_sensor = 15 [n = 7]: weight: 0.063 loss: 0.866
location_date_sensor = 16 [n = 6]: weight: 0.050 loss: 0.709
location_date_sensor = 17 [n = 15]: weight: 0.031 loss: 0.652
11% 99/915 [02:31<20:43, 1.52s/it]objective: 0.175
loss_avg: 0.823
location_date_sensor = 0 [n = 11]: weight: 0.056 loss: 0.778
location_date_sensor = 1 [n = 8]: weight: 0.025 loss: 0.569
location_date_sensor = 2 [n = 12]: weight: 0.042 loss: 0.713
location_date_sensor = 3 [n = 10]: weight: 0.047 loss: 0.775
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 0.862
location_date_sensor = 5 [n = 10]: weight: 0.039 loss: 0.616
location_date_sensor = 6 [n = 9]: weight: 0.044 loss: 0.742
location_date_sensor = 7 [n = 8]: weight: 0.098 loss: 1.004
location_date_sensor = 8 [n = 6]: weight: 0.050 loss: 0.913
location_date_sensor = 9 [n = 12]: weight: 0.099 loss: 0.864
location_date_sensor = 10 [n = 12]: weight: 0.059 loss: 0.938
location_date_sensor = 11 [n = 12]: weight: 0.063 loss: 0.878
location_date_sensor = 12 [n = 16]: weight: 0.041 loss: 0.736
location_date_sensor = 13 [n = 18]: weight: 0.076 loss: 0.930
location_date_sensor = 14 [n = 11]: weight: 0.067 loss: 0.947
location_date_sensor = 15 [n = 12]: weight: 0.062 loss: 0.961
location_date_sensor = 16 [n = 13]: weight: 0.050 loss: 0.859
location_date_sensor = 17 [n = 9]: weight: 0.031 loss: 0.623
16% 149/915 [03:48<19:29, 1.53s/it]objective: 0.165
loss_avg: 0.792
location_date_sensor = 0 [n = 8]: weight: 0.054 loss: 1.057
location_date_sensor = 1 [n = 11]: weight: 0.023 loss: 0.469
location_date_sensor = 2 [n = 12]: weight: 0.042 loss: 0.734
location_date_sensor = 3 [n = 11]: weight: 0.047 loss: 0.798
location_date_sensor = 4 [n = 7]: weight: 0.053 loss: 0.756
location_date_sensor = 5 [n = 9]: weight: 0.038 loss: 0.541
location_date_sensor = 6 [n = 12]: weight: 0.044 loss: 0.727
location_date_sensor = 7 [n = 7]: weight: 0.096 loss: 0.879
location_date_sensor = 8 [n = 13]: weight: 0.049 loss: 0.789
location_date_sensor = 9 [n = 11]: weight: 0.098 loss: 0.842
location_date_sensor = 10 [n = 8]: weight: 0.059 loss: 0.907
location_date_sensor = 11 [n = 20]: weight: 0.066 loss: 0.910
location_date_sensor = 12 [n = 11]: weight: 0.042 loss: 0.752
location_date_sensor = 13 [n = 7]: weight: 0.077 loss: 1.037
location_date_sensor = 14 [n = 7]: weight: 0.067 loss: 0.909
location_date_sensor = 15 [n = 12]: weight: 0.064 loss: 0.926
location_date_sensor = 16 [n = 20]: weight: 0.052 loss: 0.766
location_date_sensor = 17 [n = 14]: weight: 0.031 loss: 0.631
22% 199/915 [05:04<18:12, 1.53s/it]objective: 0.182
loss_avg: 0.817
location_date_sensor = 0 [n = 5]: weight: 0.055 loss: 1.292
location_date_sensor = 1 [n = 13]: weight: 0.023 loss: 0.464
location_date_sensor = 2 [n = 8]: weight: 0.042 loss: 0.735
location_date_sensor = 3 [n = 8]: weight: 0.047 loss: 0.763
location_date_sensor = 4 [n = 11]: weight: 0.051 loss: 0.836
location_date_sensor = 5 [n = 14]: weight: 0.038 loss: 0.612
location_date_sensor = 6 [n = 9]: weight: 0.043 loss: 0.775
location_date_sensor = 7 [n = 12]: weight: 0.096 loss: 0.934
location_date_sensor = 8 [n = 11]: weight: 0.049 loss: 0.826
location_date_sensor = 9 [n = 13]: weight: 0.099 loss: 0.821
location_date_sensor = 10 [n = 12]: weight: 0.059 loss: 0.997
location_date_sensor = 11 [n = 10]: weight: 0.068 loss: 0.931
location_date_sensor = 12 [n = 5]: weight: 0.040 loss: 0.688
location_date_sensor = 13 [n = 9]: weight: 0.077 loss: 0.881
location_date_sensor = 14 [n = 12]: weight: 0.066 loss: 0.895
location_date_sensor = 15 [n = 19]: weight: 0.067 loss: 0.879
location_date_sensor = 16 [n = 18]: weight: 0.054 loss: 0.895
location_date_sensor = 17 [n = 11]: weight: 0.030 loss: 0.629
27% 249/915 [06:20<16:55, 1.53s/it]objective: 0.160
loss_avg: 0.797
location_date_sensor = 0 [n = 5]: weight: 0.053 loss: 0.687
location_date_sensor = 1 [n = 12]: weight: 0.022 loss: 0.536
location_date_sensor = 2 [n = 16]: weight: 0.042 loss: 0.657
location_date_sensor = 3 [n = 13]: weight: 0.046 loss: 0.813
location_date_sensor = 4 [n = 11]: weight: 0.052 loss: 0.797
location_date_sensor = 5 [n = 13]: weight: 0.038 loss: 0.735
location_date_sensor = 6 [n = 8]: weight: 0.043 loss: 0.781
location_date_sensor = 7 [n = 12]: weight: 0.096 loss: 0.973
location_date_sensor = 8 [n = 13]: weight: 0.051 loss: 0.854
location_date_sensor = 9 [n = 10]: weight: 0.101 loss: 0.893
location_date_sensor = 10 [n = 7]: weight: 0.059 loss: 0.953
location_date_sensor = 11 [n = 10]: weight: 0.067 loss: 0.925
location_date_sensor = 12 [n = 10]: weight: 0.039 loss: 0.726
location_date_sensor = 13 [n = 12]: weight: 0.076 loss: 0.909
location_date_sensor = 14 [n = 7]: weight: 0.065 loss: 0.878
location_date_sensor = 15 [n = 9]: weight: 0.067 loss: 0.939
location_date_sensor = 16 [n = 16]: weight: 0.056 loss: 0.801
location_date_sensor = 17 [n = 16]: weight: 0.031 loss: 0.664
33% 299/915 [07:37<15:40, 1.53s/it]objective: 0.181
loss_avg: 0.810
location_date_sensor = 0 [n = 12]: weight: 0.052 loss: 1.013
location_date_sensor = 1 [n = 7]: weight: 0.022 loss: 0.584
location_date_sensor = 2 [n = 8]: weight: 0.042 loss: 0.625
location_date_sensor = 3 [n = 13]: weight: 0.046 loss: 0.723
location_date_sensor = 4 [n = 13]: weight: 0.052 loss: 0.827
location_date_sensor = 5 [n = 9]: weight: 0.037 loss: 0.555
location_date_sensor = 6 [n = 8]: weight: 0.041 loss: 0.662
location_date_sensor = 7 [n = 11]: weight: 0.097 loss: 0.978
location_date_sensor = 8 [n = 16]: weight: 0.052 loss: 0.796
location_date_sensor = 9 [n = 10]: weight: 0.100 loss: 0.824
location_date_sensor = 10 [n = 13]: weight: 0.060 loss: 0.932
location_date_sensor = 11 [n = 15]: weight: 0.068 loss: 0.889
location_date_sensor = 12 [n = 11]: weight: 0.039 loss: 0.758
location_date_sensor = 13 [n = 9]: weight: 0.077 loss: 0.932
location_date_sensor = 14 [n = 11]: weight: 0.065 loss: 0.915
location_date_sensor = 15 [n = 8]: weight: 0.066 loss: 0.926
location_date_sensor = 16 [n = 18]: weight: 0.058 loss: 0.783
location_date_sensor = 17 [n = 8]: weight: 0.030 loss: 0.606
38% 349/915 [08:53<14:24, 1.53s/it]objective: 0.174
loss_avg: 0.807
location_date_sensor = 0 [n = 13]: weight: 0.053 loss: 1.013
location_date_sensor = 1 [n = 16]: weight: 0.020 loss: 0.523
location_date_sensor = 2 [n = 9]: weight: 0.039 loss: 0.731
location_date_sensor = 3 [n = 7]: weight: 0.045 loss: 0.698
location_date_sensor = 4 [n = 10]: weight: 0.052 loss: 0.805
location_date_sensor = 5 [n = 11]: weight: 0.036 loss: 0.607
location_date_sensor = 6 [n = 8]: weight: 0.039 loss: 0.766
location_date_sensor = 7 [n = 15]: weight: 0.101 loss: 0.902
location_date_sensor = 8 [n = 12]: weight: 0.053 loss: 0.826
location_date_sensor = 9 [n = 9]: weight: 0.101 loss: 0.943
location_date_sensor = 10 [n = 9]: weight: 0.060 loss: 0.958
location_date_sensor = 11 [n = 11]: weight: 0.070 loss: 0.865
location_date_sensor = 12 [n = 16]: weight: 0.039 loss: 0.780
location_date_sensor = 13 [n = 12]: weight: 0.077 loss: 0.998
location_date_sensor = 14 [n = 11]: weight: 0.065 loss: 0.905
location_date_sensor = 15 [n = 11]: weight: 0.067 loss: 0.865
location_date_sensor = 16 [n = 8]: weight: 0.057 loss: 0.794
location_date_sensor = 17 [n = 12]: weight: 0.029 loss: 0.597
44% 399/915 [10:09<13:08, 1.53s/it]objective: 0.180
loss_avg: 0.790
location_date_sensor = 0 [n = 15]: weight: 0.054 loss: 0.784
location_date_sensor = 1 [n = 4]: weight: 0.019 loss: 0.588
location_date_sensor = 2 [n = 10]: weight: 0.039 loss: 0.651
location_date_sensor = 3 [n = 12]: weight: 0.044 loss: 0.684
location_date_sensor = 4 [n = 7]: weight: 0.050 loss: 0.790
location_date_sensor = 5 [n = 7]: weight: 0.034 loss: 0.703
location_date_sensor = 6 [n = 10]: weight: 0.039 loss: 0.734
location_date_sensor = 7 [n = 19]: weight: 0.105 loss: 0.808
location_date_sensor = 8 [n = 9]: weight: 0.052 loss: 0.765
location_date_sensor = 9 [n = 12]: weight: 0.100 loss: 0.808
location_date_sensor = 10 [n = 16]: weight: 0.062 loss: 0.952
location_date_sensor = 11 [n = 9]: weight: 0.069 loss: 0.840
location_date_sensor = 12 [n = 12]: weight: 0.039 loss: 0.726
location_date_sensor = 13 [n = 16]: weight: 0.079 loss: 0.848
location_date_sensor = 14 [n = 6]: weight: 0.064 loss: 0.943
location_date_sensor = 15 [n = 10]: weight: 0.067 loss: 0.871
location_date_sensor = 16 [n = 15]: weight: 0.058 loss: 0.861
location_date_sensor = 17 [n = 11]: weight: 0.028 loss: 0.647
49% 449/915 [11:26<11:52, 1.53s/it]objective: 0.153
loss_avg: 0.754
location_date_sensor = 0 [n = 5]: weight: 0.053 loss: 0.721
location_date_sensor = 1 [n = 13]: weight: 0.019 loss: 0.550
location_date_sensor = 2 [n = 12]: weight: 0.038 loss: 0.669
location_date_sensor = 3 [n = 8]: weight: 0.043 loss: 0.666
location_date_sensor = 4 [n = 12]: weight: 0.050 loss: 0.781
location_date_sensor = 5 [n = 13]: weight: 0.034 loss: 0.659
location_date_sensor = 6 [n = 8]: weight: 0.038 loss: 0.853
location_date_sensor = 7 [n = 10]: weight: 0.108 loss: 0.752
location_date_sensor = 8 [n = 11]: weight: 0.052 loss: 0.712
location_date_sensor = 9 [n = 12]: weight: 0.100 loss: 0.892
location_date_sensor = 10 [n = 6]: weight: 0.064 loss: 0.967
location_date_sensor = 11 [n = 9]: weight: 0.069 loss: 0.856
location_date_sensor = 12 [n = 15]: weight: 0.040 loss: 0.764
location_date_sensor = 13 [n = 11]: weight: 0.080 loss: 0.879
location_date_sensor = 14 [n = 14]: weight: 0.065 loss: 0.935
location_date_sensor = 15 [n = 10]: weight: 0.065 loss: 0.854
location_date_sensor = 16 [n = 10]: weight: 0.059 loss: 0.713
location_date_sensor = 17 [n = 21]: weight: 0.029 loss: 0.590
55% 499/915 [12:42<10:33, 1.52s/it]objective: 0.165
loss_avg: 0.776
location_date_sensor = 0 [n = 16]: weight: 0.052 loss: 0.817
location_date_sensor = 1 [n = 11]: weight: 0.019 loss: 0.520
location_date_sensor = 2 [n = 9]: weight: 0.038 loss: 0.626
location_date_sensor = 3 [n = 9]: weight: 0.041 loss: 0.662
location_date_sensor = 4 [n = 17]: weight: 0.052 loss: 0.785
location_date_sensor = 5 [n = 12]: weight: 0.034 loss: 0.593
location_date_sensor = 6 [n = 13]: weight: 0.037 loss: 0.715
location_date_sensor = 7 [n = 6]: weight: 0.105 loss: 0.873
location_date_sensor = 8 [n = 8]: weight: 0.051 loss: 0.707
location_date_sensor = 9 [n = 15]: weight: 0.103 loss: 0.826
location_date_sensor = 10 [n = 12]: weight: 0.062 loss: 0.885
location_date_sensor = 11 [n = 11]: weight: 0.069 loss: 0.860
location_date_sensor = 12 [n = 5]: weight: 0.039 loss: 0.761
location_date_sensor = 13 [n = 9]: weight: 0.081 loss: 0.882
location_date_sensor = 14 [n = 18]: weight: 0.069 loss: 0.907
location_date_sensor = 15 [n = 13]: weight: 0.065 loss: 0.907
location_date_sensor = 16 [n = 9]: weight: 0.059 loss: 0.839
location_date_sensor = 17 [n = 7]: weight: 0.028 loss: 0.601
60% 549/915 [13:58<09:18, 1.53s/it]objective: 0.177
loss_avg: 0.789
location_date_sensor = 0 [n = 15]: weight: 0.054 loss: 0.934
location_date_sensor = 1 [n = 8]: weight: 0.018 loss: 0.592
location_date_sensor = 2 [n = 12]: weight: 0.036 loss: 0.687
location_date_sensor = 3 [n = 12]: weight: 0.040 loss: 0.677
location_date_sensor = 4 [n = 10]: weight: 0.053 loss: 0.804
location_date_sensor = 5 [n = 8]: weight: 0.033 loss: 0.585
location_date_sensor = 6 [n = 10]: weight: 0.037 loss: 0.745
location_date_sensor = 7 [n = 8]: weight: 0.102 loss: 0.862
location_date_sensor = 8 [n = 8]: weight: 0.049 loss: 0.795
location_date_sensor = 9 [n = 12]: weight: 0.103 loss: 0.827
location_date_sensor = 10 [n = 13]: weight: 0.063 loss: 0.987
location_date_sensor = 11 [n = 13]: weight: 0.070 loss: 0.789
location_date_sensor = 12 [n = 10]: weight: 0.039 loss: 0.722
location_date_sensor = 13 [n = 17]: weight: 0.084 loss: 0.855
location_date_sensor = 14 [n = 11]: weight: 0.071 loss: 0.825
location_date_sensor = 15 [n = 10]: weight: 0.065 loss: 0.905
location_date_sensor = 16 [n = 8]: weight: 0.058 loss: 0.875
location_date_sensor = 17 [n = 15]: weight: 0.028 loss: 0.626
65% 599/915 [15:15<08:00, 1.52s/it]objective: 0.165
loss_avg: 0.777
location_date_sensor = 0 [n = 13]: weight: 0.056 loss: 0.793
location_date_sensor = 1 [n = 9]: weight: 0.017 loss: 0.570
location_date_sensor = 2 [n = 8]: weight: 0.036 loss: 0.677
location_date_sensor = 3 [n = 10]: weight: 0.040 loss: 0.706
location_date_sensor = 4 [n = 13]: weight: 0.053 loss: 0.737
location_date_sensor = 5 [n = 13]: weight: 0.032 loss: 0.818
location_date_sensor = 6 [n = 18]: weight: 0.037 loss: 0.786
location_date_sensor = 7 [n = 14]: weight: 0.104 loss: 0.854
location_date_sensor = 8 [n = 5]: weight: 0.047 loss: 0.779
location_date_sensor = 9 [n = 12]: weight: 0.104 loss: 0.762
location_date_sensor = 10 [n = 5]: weight: 0.063 loss: 0.905
location_date_sensor = 11 [n = 13]: weight: 0.071 loss: 0.786
location_date_sensor = 12 [n = 10]: weight: 0.038 loss: 0.755
location_date_sensor = 13 [n = 16]: weight: 0.087 loss: 0.877
location_date_sensor = 14 [n = 7]: weight: 0.069 loss: 0.908
location_date_sensor = 15 [n = 11]: weight: 0.065 loss: 0.935
location_date_sensor = 16 [n = 11]: weight: 0.056 loss: 0.715
location_date_sensor = 17 [n = 12]: weight: 0.028 loss: 0.614
71% 649/915 [16:31<06:46, 1.53s/it]objective: 0.167
loss_avg: 0.754
location_date_sensor = 0 [n = 17]: weight: 0.056 loss: 0.838
location_date_sensor = 1 [n = 13]: weight: 0.017 loss: 0.555
location_date_sensor = 2 [n = 15]: weight: 0.035 loss: 0.663
location_date_sensor = 3 [n = 9]: weight: 0.039 loss: 0.707
location_date_sensor = 4 [n = 10]: weight: 0.052 loss: 0.763
location_date_sensor = 5 [n = 7]: weight: 0.032 loss: 0.542
location_date_sensor = 6 [n = 11]: weight: 0.038 loss: 0.690
location_date_sensor = 7 [n = 14]: weight: 0.106 loss: 0.853
location_date_sensor = 8 [n = 12]: weight: 0.047 loss: 0.761
location_date_sensor = 9 [n = 12]: weight: 0.104 loss: 0.800
location_date_sensor = 10 [n = 13]: weight: 0.063 loss: 0.952
location_date_sensor = 11 [n = 3]: weight: 0.070 loss: 0.708
location_date_sensor = 12 [n = 10]: weight: 0.038 loss: 0.737
location_date_sensor = 13 [n = 11]: weight: 0.089 loss: 0.864
location_date_sensor = 14 [n = 5]: weight: 0.068 loss: 0.867
location_date_sensor = 15 [n = 11]: weight: 0.066 loss: 0.836
location_date_sensor = 16 [n = 13]: weight: 0.057 loss: 0.724
location_date_sensor = 17 [n = 14]: weight: 0.028 loss: 0.643
76% 699/915 [17:47<05:29, 1.53s/it]objective: 0.169
loss_avg: 0.769
location_date_sensor = 0 [n = 19]: weight: 0.060 loss: 0.924
location_date_sensor = 1 [n = 9]: weight: 0.016 loss: 0.560
location_date_sensor = 2 [n = 10]: weight: 0.035 loss: 0.652
location_date_sensor = 3 [n = 11]: weight: 0.039 loss: 0.662
location_date_sensor = 4 [n = 11]: weight: 0.051 loss: 0.783
location_date_sensor = 5 [n = 10]: weight: 0.031 loss: 0.558
location_date_sensor = 6 [n = 8]: weight: 0.038 loss: 0.798
location_date_sensor = 7 [n = 14]: weight: 0.108 loss: 0.820
location_date_sensor = 8 [n = 10]: weight: 0.047 loss: 0.718
location_date_sensor = 9 [n = 7]: weight: 0.102 loss: 0.804
location_date_sensor = 10 [n = 10]: weight: 0.064 loss: 0.885
location_date_sensor = 11 [n = 10]: weight: 0.066 loss: 0.809
location_date_sensor = 12 [n = 12]: weight: 0.037 loss: 0.642
location_date_sensor = 13 [n = 15]: weight: 0.093 loss: 0.885
location_date_sensor = 14 [n = 9]: weight: 0.067 loss: 0.977
location_date_sensor = 15 [n = 10]: weight: 0.065 loss: 0.821
location_date_sensor = 16 [n = 15]: weight: 0.056 loss: 0.731
location_date_sensor = 17 [n = 10]: weight: 0.027 loss: 0.659
82% 749/915 [19:04<04:14, 1.53s/it]objective: 0.172
loss_avg: 0.752
location_date_sensor = 0 [n = 9]: weight: 0.063 loss: 0.789
location_date_sensor = 1 [n = 11]: weight: 0.016 loss: 0.559
location_date_sensor = 2 [n = 12]: weight: 0.034 loss: 0.697
location_date_sensor = 3 [n = 5]: weight: 0.036 loss: 0.713
location_date_sensor = 4 [n = 11]: weight: 0.051 loss: 0.711
location_date_sensor = 5 [n = 9]: weight: 0.029 loss: 0.600
location_date_sensor = 6 [n = 15]: weight: 0.037 loss: 0.707
location_date_sensor = 7 [n = 12]: weight: 0.112 loss: 0.799
location_date_sensor = 8 [n = 18]: weight: 0.047 loss: 0.751
location_date_sensor = 9 [n = 12]: weight: 0.102 loss: 0.773
location_date_sensor = 10 [n = 13]: weight: 0.065 loss: 0.866
location_date_sensor = 11 [n = 12]: weight: 0.066 loss: 0.805
location_date_sensor = 12 [n = 9]: weight: 0.036 loss: 0.778
location_date_sensor = 13 [n = 16]: weight: 0.096 loss: 0.860
location_date_sensor = 14 [n = 9]: weight: 0.066 loss: 0.929
location_date_sensor = 15 [n = 7]: weight: 0.064 loss: 0.880
location_date_sensor = 16 [n = 14]: weight: 0.056 loss: 0.674
location_date_sensor = 17 [n = 6]: weight: 0.026 loss: 0.575
87% 799/915 [20:20<02:57, 1.53s/it]objective: 0.152
loss_avg: 0.718
location_date_sensor = 0 [n = 10]: weight: 0.062 loss: 0.692
location_date_sensor = 1 [n = 12]: weight: 0.016 loss: 0.485
location_date_sensor = 2 [n = 8]: weight: 0.034 loss: 0.637
location_date_sensor = 3 [n = 12]: weight: 0.036 loss: 0.616
location_date_sensor = 4 [n = 15]: weight: 0.052 loss: 0.718
location_date_sensor = 5 [n = 11]: weight: 0.028 loss: 0.580
location_date_sensor = 6 [n = 14]: weight: 0.038 loss: 0.744
location_date_sensor = 7 [n = 10]: weight: 0.112 loss: 0.700
location_date_sensor = 8 [n = 9]: weight: 0.048 loss: 0.740
location_date_sensor = 9 [n = 14]: weight: 0.102 loss: 0.761
location_date_sensor = 10 [n = 10]: weight: 0.066 loss: 0.852
location_date_sensor = 11 [n = 11]: weight: 0.066 loss: 0.739
location_date_sensor = 12 [n = 10]: weight: 0.036 loss: 0.743
location_date_sensor = 13 [n = 8]: weight: 0.096 loss: 0.844
location_date_sensor = 14 [n = 16]: weight: 0.066 loss: 0.877
location_date_sensor = 15 [n = 13]: weight: 0.064 loss: 0.842
location_date_sensor = 16 [n = 4]: weight: 0.054 loss: 0.692
location_date_sensor = 17 [n = 13]: weight: 0.026 loss: 0.613
93% 849/915 [21:36<01:40, 1.53s/it]objective: 0.171
loss_avg: 0.767
location_date_sensor = 0 [n = 13]: weight: 0.062 loss: 0.694
location_date_sensor = 1 [n = 9]: weight: 0.015 loss: 0.506
location_date_sensor = 2 [n = 2]: weight: 0.032 loss: 0.655
location_date_sensor = 3 [n = 11]: weight: 0.035 loss: 0.643
location_date_sensor = 4 [n = 10]: weight: 0.051 loss: 0.744
location_date_sensor = 5 [n = 12]: weight: 0.028 loss: 0.611
location_date_sensor = 6 [n = 10]: weight: 0.039 loss: 0.815
location_date_sensor = 7 [n = 11]: weight: 0.110 loss: 0.772
location_date_sensor = 8 [n = 10]: weight: 0.047 loss: 0.706
location_date_sensor = 9 [n = 10]: weight: 0.103 loss: 0.827
location_date_sensor = 10 [n = 12]: weight: 0.066 loss: 0.894
location_date_sensor = 11 [n = 15]: weight: 0.066 loss: 0.764
location_date_sensor = 12 [n = 12]: weight: 0.036 loss: 0.728
location_date_sensor = 13 [n = 13]: weight: 0.098 loss: 0.926
location_date_sensor = 14 [n = 14]: weight: 0.069 loss: 0.910
location_date_sensor = 15 [n = 17]: weight: 0.067 loss: 0.914
location_date_sensor = 16 [n = 9]: weight: 0.053 loss: 0.847
location_date_sensor = 17 [n = 10]: weight: 0.025 loss: 0.592
98% 899/915 [22:53<00:24, 1.53s/it]objective: 0.171
loss_avg: 0.763
location_date_sensor = 0 [n = 12]: weight: 0.062 loss: 0.832
location_date_sensor = 1 [n = 11]: weight: 0.014 loss: 0.469
location_date_sensor = 2 [n = 14]: weight: 0.031 loss: 0.653
location_date_sensor = 3 [n = 8]: weight: 0.035 loss: 0.690
location_date_sensor = 4 [n = 16]: weight: 0.052 loss: 0.747
location_date_sensor = 5 [n = 6]: weight: 0.028 loss: 0.608
location_date_sensor = 6 [n = 10]: weight: 0.039 loss: 0.847
location_date_sensor = 7 [n = 18]: weight: 0.113 loss: 0.805
location_date_sensor = 8 [n = 9]: weight: 0.046 loss: 0.801
location_date_sensor = 9 [n = 9]: weight: 0.100 loss: 0.812
location_date_sensor = 10 [n = 12]: weight: 0.067 loss: 0.849
location_date_sensor = 11 [n = 9]: weight: 0.065 loss: 0.732
location_date_sensor = 12 [n = 14]: weight: 0.036 loss: 0.779
location_date_sensor = 13 [n = 12]: weight: 0.101 loss: 0.955
location_date_sensor = 14 [n = 6]: weight: 0.068 loss: 0.844
location_date_sensor = 15 [n = 12]: weight: 0.070 loss: 0.904
location_date_sensor = 16 [n = 10]: weight: 0.052 loss: 0.725
location_date_sensor = 17 [n = 12]: weight: 0.025 loss: 0.611
100% 915/915 [23:16<00:00, 1.53s/it]
objective: 0.123
loss_avg: 0.684
location_date_sensor = 0 [n = 3]: weight: 0.062 loss: 0.680
location_date_sensor = 1 [n = 6]: weight: 0.014 loss: 0.499
location_date_sensor = 2 [n = 5]: weight: 0.031 loss: 0.705
location_date_sensor = 3 [n = 4]: weight: 0.034 loss: 0.653
location_date_sensor = 4 [n = 3]: weight: 0.052 loss: 0.794
location_date_sensor = 5 [n = 6]: weight: 0.027 loss: 0.547
location_date_sensor = 6 [n = 2]: weight: 0.039 loss: 0.693
location_date_sensor = 7 [n = 1]: weight: 0.116 loss: 0.649
location_date_sensor = 8 [n = 1]: weight: 0.046 loss: 0.695
location_date_sensor = 9 [n = 4]: weight: 0.099 loss: 0.760
location_date_sensor = 10 [n = 3]: weight: 0.068 loss: 0.917
location_date_sensor = 11 [n = 3]: weight: 0.065 loss: 0.774
location_date_sensor = 12 [n = 4]: weight: 0.037 loss: 0.707
location_date_sensor = 13 [n = 1]: weight: 0.101 loss: 0.844
location_date_sensor = 14 [n = 2]: weight: 0.066 loss: 0.947
location_date_sensor = 15 [n = 3]: weight: 0.071 loss: 0.788
location_date_sensor = 16 [n = 2]: weight: 0.051 loss: 0.635
location_date_sensor = 17 [n = 4]: weight: 0.024 loss: 0.533
Epoch eval:
Average detection_acc: 0.722
location_date_sensor = 0 [n = 211]: detection_acc = 0.702
location_date_sensor = 1 [n = 194]: detection_acc = 0.878
location_date_sensor = 2 [n = 190]: detection_acc = 0.848
location_date_sensor = 3 [n = 188]: detection_acc = 0.768
location_date_sensor = 4 [n = 212]: detection_acc = 0.745
location_date_sensor = 5 [n = 192]: detection_acc = 0.771
location_date_sensor = 6 [n = 191]: detection_acc = 0.746
location_date_sensor = 7 [n = 211]: detection_acc = 0.657
location_date_sensor = 8 [n = 191]: detection_acc = 0.706
location_date_sensor = 9 [n = 213]: detection_acc = 0.593
location_date_sensor = 10 [n = 194]: detection_acc = 0.659
location_date_sensor = 11 [n = 211]: detection_acc = 0.608
location_date_sensor = 12 [n = 204]: detection_acc = 0.725
location_date_sensor = 13 [n = 225]: detection_acc = 0.707
location_date_sensor = 14 [n = 186]: detection_acc = 0.713
location_date_sensor = 15 [n = 205]: detection_acc = 0.754
location_date_sensor = 16 [n = 219]: detection_acc = 0.626
location_date_sensor = 17 [n = 220]: detection_acc = 0.818
Worst-group detection_acc: 0.593
Validation:
100% 348/348 [03:19<00:00, 1.75it/s]
objective: 0.000
loss_avg: 1.000
location_date_sensor = 18 [n = 12]: weight: 0.000 loss: 1.000
location_date_sensor = 19 [n = 49]: weight: 0.000 loss: 1.000
location_date_sensor = 20 [n = 254]: weight: 0.000 loss: 1.000
location_date_sensor = 21 [n = 216]: weight: 0.000 loss: 1.000
location_date_sensor = 22 [n = 89]: weight: 0.000 loss: 1.000
location_date_sensor = 23 [n = 11]: weight: 0.000 loss: 1.000
location_date_sensor = 24 [n = 4]: weight: 0.000 loss: 1.000
location_date_sensor = 25 [n = 7]: weight: 0.000 loss: 1.000
location_date_sensor = 26 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 27 [n = 17]: weight: 0.000 loss: 1.000
location_date_sensor = 28 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 29 [n = 8]: weight: 0.000 loss: 1.000
location_date_sensor = 30 [n = 43]: weight: 0.000 loss: 1.000
location_date_sensor = 31 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 32 [n = 51]: weight: 0.000 loss: 1.000
location_date_sensor = 33 [n = 50]: weight: 0.000 loss: 1.000
location_date_sensor = 34 [n = 28]: weight: 0.000 loss: 1.000
location_date_sensor = 35 [n = 75]: weight: 0.000 loss: 1.000
location_date_sensor = 36 [n = 56]: weight: 0.000 loss: 1.000
location_date_sensor = 37 [n = 34]: weight: 0.000 loss: 1.000
location_date_sensor = 38 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 39 [n = 13]: weight: 0.000 loss: 1.000
location_date_sensor = 40 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 41 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 42 [n = 57]: weight: 0.000 loss: 1.000
location_date_sensor = 43 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 44 [n = 33]: weight: 0.000 loss: 1.000
location_date_sensor = 45 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 46 [n = 30]: weight: 0.000 loss: 1.000
Epoch eval:
Average detection_acc: 0.009
location_date_sensor = 18 [n = 12]: detection_acc = 0.000
location_date_sensor = 19 [n = 49]: detection_acc = 0.020
location_date_sensor = 20 [n = 254]: detection_acc = 0.000
location_date_sensor = 21 [n = 216]: detection_acc = 0.000
location_date_sensor = 22 [n = 89]: detection_acc = 0.000
location_date_sensor = 23 [n = 11]: detection_acc = 0.000
location_date_sensor = 24 [n = 4]: detection_acc = 0.500
location_date_sensor = 25 [n = 7]: detection_acc = 0.286
location_date_sensor = 26 [n = 14]: detection_acc = 0.000
location_date_sensor = 27 [n = 17]: detection_acc = 0.059
location_date_sensor = 28 [n = 14]: detection_acc = 0.000
location_date_sensor = 29 [n = 8]: detection_acc = 0.000
location_date_sensor = 30 [n = 43]: detection_acc = 0.000
location_date_sensor = 31 [n = 55]: detection_acc = 0.000
location_date_sensor = 32 [n = 51]: detection_acc = 0.000
location_date_sensor = 33 [n = 50]: detection_acc = 0.000
location_date_sensor = 34 [n = 28]: detection_acc = 0.000
location_date_sensor = 35 [n = 75]: detection_acc = 0.040
location_date_sensor = 36 [n = 56]: detection_acc = 0.000
location_date_sensor = 37 [n = 34]: detection_acc = 0.000
location_date_sensor = 38 [n = 55]: detection_acc = 0.000
location_date_sensor = 39 [n = 13]: detection_acc = 0.000
location_date_sensor = 40 [n = 19]: detection_acc = 0.000
location_date_sensor = 41 [n = 19]: detection_acc = 0.000
location_date_sensor = 42 [n = 57]: detection_acc = 0.000
location_date_sensor = 43 [n = 39]: detection_acc = 0.000
location_date_sensor = 44 [n = 33]: detection_acc = 0.000
location_date_sensor = 45 [n = 39]: detection_acc = 0.026
location_date_sensor = 46 [n = 30]: detection_acc = 0.100
Worst-group detection_acc: 0.000
Validation detection_acc_avg: 0.009
Epoch 1 has the best validation performance so far.
100% 365/365 [03:28<00:00, 1.75it/s]
Epoch [2]:
Train:
5% 49/915 [01:15<22:02, 1.53s/it]objective: 0.159
loss_avg: 0.736
location_date_sensor = 0 [n = 8]: weight: 0.061 loss: 0.693
location_date_sensor = 1 [n = 10]: weight: 0.014 loss: 0.549
location_date_sensor = 2 [n = 14]: weight: 0.031 loss: 0.729
location_date_sensor = 3 [n = 12]: weight: 0.034 loss: 0.600
location_date_sensor = 4 [n = 14]: weight: 0.052 loss: 0.702
location_date_sensor = 5 [n = 10]: weight: 0.026 loss: 0.470
location_date_sensor = 6 [n = 8]: weight: 0.038 loss: 0.771
location_date_sensor = 7 [n = 8]: weight: 0.113 loss: 0.781
location_date_sensor = 8 [n = 11]: weight: 0.046 loss: 0.760
location_date_sensor = 9 [n = 12]: weight: 0.100 loss: 0.770
location_date_sensor = 10 [n = 13]: weight: 0.069 loss: 0.857
location_date_sensor = 11 [n = 8]: weight: 0.064 loss: 0.767
location_date_sensor = 12 [n = 11]: weight: 0.037 loss: 0.699
location_date_sensor = 13 [n = 13]: weight: 0.103 loss: 0.874
location_date_sensor = 14 [n = 10]: weight: 0.065 loss: 0.882
location_date_sensor = 15 [n = 8]: weight: 0.071 loss: 0.927
location_date_sensor = 16 [n = 20]: weight: 0.053 loss: 0.760
location_date_sensor = 17 [n = 10]: weight: 0.024 loss: 0.644
11% 99/915 [02:31<20:46, 1.53s/it]objective: 0.154
loss_avg: 0.719
location_date_sensor = 0 [n = 12]: weight: 0.061 loss: 0.714
location_date_sensor = 1 [n = 18]: weight: 0.014 loss: 0.534
location_date_sensor = 2 [n = 7]: weight: 0.031 loss: 0.634
location_date_sensor = 3 [n = 17]: weight: 0.034 loss: 0.623
location_date_sensor = 4 [n = 6]: weight: 0.052 loss: 0.773
location_date_sensor = 5 [n = 11]: weight: 0.026 loss: 0.551
location_date_sensor = 6 [n = 7]: weight: 0.037 loss: 0.647
location_date_sensor = 7 [n = 12]: weight: 0.113 loss: 0.759
location_date_sensor = 8 [n = 10]: weight: 0.045 loss: 0.671
location_date_sensor = 9 [n = 10]: weight: 0.102 loss: 0.786
location_date_sensor = 10 [n = 11]: weight: 0.070 loss: 0.941
location_date_sensor = 11 [n = 19]: weight: 0.064 loss: 0.686
location_date_sensor = 12 [n = 8]: weight: 0.036 loss: 0.723
location_date_sensor = 13 [n = 11]: weight: 0.104 loss: 0.875
location_date_sensor = 14 [n = 6]: weight: 0.066 loss: 0.902
location_date_sensor = 15 [n = 12]: weight: 0.071 loss: 0.859
location_date_sensor = 16 [n = 15]: weight: 0.055 loss: 0.772
location_date_sensor = 17 [n = 8]: weight: 0.024 loss: 0.692
16% 149/915 [03:47<19:29, 1.53s/it]objective: 0.160
loss_avg: 0.739
location_date_sensor = 0 [n = 10]: weight: 0.061 loss: 0.808
location_date_sensor = 1 [n = 9]: weight: 0.014 loss: 0.543
location_date_sensor = 2 [n = 13]: weight: 0.030 loss: 0.691
location_date_sensor = 3 [n = 8]: weight: 0.033 loss: 0.668
location_date_sensor = 4 [n = 10]: weight: 0.051 loss: 0.729
location_date_sensor = 5 [n = 13]: weight: 0.026 loss: 0.582
location_date_sensor = 6 [n = 4]: weight: 0.036 loss: 0.785
location_date_sensor = 7 [n = 11]: weight: 0.111 loss: 0.711
location_date_sensor = 8 [n = 17]: weight: 0.046 loss: 0.701
location_date_sensor = 9 [n = 17]: weight: 0.104 loss: 0.787
location_date_sensor = 10 [n = 14]: weight: 0.072 loss: 0.898
location_date_sensor = 11 [n = 15]: weight: 0.066 loss: 0.677
location_date_sensor = 12 [n = 5]: weight: 0.035 loss: 0.812
location_date_sensor = 13 [n = 9]: weight: 0.103 loss: 0.802
location_date_sensor = 14 [n = 12]: weight: 0.066 loss: 0.862
location_date_sensor = 15 [n = 15]: weight: 0.072 loss: 0.887
location_date_sensor = 16 [n = 10]: weight: 0.054 loss: 0.692
location_date_sensor = 17 [n = 8]: weight: 0.023 loss: 0.613
22% 199/915 [05:04<18:11, 1.52s/it]objective: 0.164
loss_avg: 0.734
location_date_sensor = 0 [n = 8]: weight: 0.060 loss: 0.643
location_date_sensor = 1 [n = 11]: weight: 0.014 loss: 0.587
location_date_sensor = 2 [n = 10]: weight: 0.030 loss: 0.613
location_date_sensor = 3 [n = 8]: weight: 0.032 loss: 0.584
location_date_sensor = 4 [n = 14]: weight: 0.051 loss: 0.765
location_date_sensor = 5 [n = 12]: weight: 0.025 loss: 0.656
location_date_sensor = 6 [n = 9]: weight: 0.035 loss: 0.693
location_date_sensor = 7 [n = 16]: weight: 0.113 loss: 0.844
location_date_sensor = 8 [n = 9]: weight: 0.046 loss: 0.716
location_date_sensor = 9 [n = 10]: weight: 0.107 loss: 0.756
location_date_sensor = 10 [n = 11]: weight: 0.073 loss: 0.809
location_date_sensor = 11 [n = 12]: weight: 0.066 loss: 0.708
location_date_sensor = 12 [n = 11]: weight: 0.034 loss: 0.781
location_date_sensor = 13 [n = 9]: weight: 0.102 loss: 0.851
location_date_sensor = 14 [n = 13]: weight: 0.066 loss: 0.858
location_date_sensor = 15 [n = 8]: weight: 0.074 loss: 0.824
location_date_sensor = 16 [n = 17]: weight: 0.054 loss: 0.775
location_date_sensor = 17 [n = 12]: weight: 0.022 loss: 0.624
27% 249/915 [06:20<17:00, 1.53s/it]objective: 0.161
loss_avg: 0.741
location_date_sensor = 0 [n = 13]: weight: 0.060 loss: 0.766
location_date_sensor = 1 [n = 12]: weight: 0.013 loss: 0.561
location_date_sensor = 2 [n = 15]: weight: 0.030 loss: 0.712
location_date_sensor = 3 [n = 12]: weight: 0.032 loss: 0.642
location_date_sensor = 4 [n = 12]: weight: 0.053 loss: 0.754
location_date_sensor = 5 [n = 7]: weight: 0.025 loss: 0.724
location_date_sensor = 6 [n = 10]: weight: 0.034 loss: 0.792
location_date_sensor = 7 [n = 12]: weight: 0.117 loss: 0.706
location_date_sensor = 8 [n = 21]: weight: 0.047 loss: 0.705
location_date_sensor = 9 [n = 7]: weight: 0.102 loss: 0.720
location_date_sensor = 10 [n = 11]: weight: 0.074 loss: 0.901
location_date_sensor = 11 [n = 4]: weight: 0.064 loss: 0.777
location_date_sensor = 12 [n = 9]: weight: 0.034 loss: 0.744
location_date_sensor = 13 [n = 11]: weight: 0.102 loss: 0.845
location_date_sensor = 14 [n = 15]: weight: 0.068 loss: 0.843
location_date_sensor = 15 [n = 10]: weight: 0.074 loss: 0.936
location_date_sensor = 16 [n = 5]: weight: 0.054 loss: 0.694
location_date_sensor = 17 [n = 14]: weight: 0.022 loss: 0.598
33% 299/915 [07:36<15:40, 1.53s/it]objective: 0.159
loss_avg: 0.732
location_date_sensor = 0 [n = 10]: weight: 0.059 loss: 0.648
location_date_sensor = 1 [n = 9]: weight: 0.013 loss: 0.550
location_date_sensor = 2 [n = 12]: weight: 0.030 loss: 0.676
location_date_sensor = 3 [n = 8]: weight: 0.031 loss: 0.629
location_date_sensor = 4 [n = 15]: weight: 0.053 loss: 0.716
location_date_sensor = 5 [n = 8]: weight: 0.024 loss: 0.810
location_date_sensor = 6 [n = 7]: weight: 0.034 loss: 0.768
location_date_sensor = 7 [n = 16]: weight: 0.117 loss: 0.770
location_date_sensor = 8 [n = 9]: weight: 0.047 loss: 0.741
location_date_sensor = 9 [n = 13]: weight: 0.103 loss: 0.716
location_date_sensor = 10 [n = 17]: weight: 0.077 loss: 0.930
location_date_sensor = 11 [n = 14]: weight: 0.063 loss: 0.659
location_date_sensor = 12 [n = 12]: weight: 0.034 loss: 0.670
location_date_sensor = 13 [n = 6]: weight: 0.100 loss: 0.807
location_date_sensor = 14 [n = 5]: weight: 0.068 loss: 0.928
location_date_sensor = 15 [n = 8]: weight: 0.073 loss: 0.864
location_date_sensor = 16 [n = 18]: weight: 0.054 loss: 0.757
location_date_sensor = 17 [n = 13]: weight: 0.022 loss: 0.597
38% 349/915 [08:53<14:20, 1.52s/it]objective: 0.150
loss_avg: 0.742
location_date_sensor = 0 [n = 11]: weight: 0.059 loss: 0.903
location_date_sensor = 1 [n = 13]: weight: 0.013 loss: 0.568
location_date_sensor = 2 [n = 13]: weight: 0.030 loss: 0.622
location_date_sensor = 3 [n = 17]: weight: 0.031 loss: 0.623
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 0.696
location_date_sensor = 5 [n = 9]: weight: 0.024 loss: 0.508
location_date_sensor = 6 [n = 12]: weight: 0.033 loss: 0.816
location_date_sensor = 7 [n = 6]: weight: 0.117 loss: 0.763
location_date_sensor = 8 [n = 5]: weight: 0.046 loss: 0.728
location_date_sensor = 9 [n = 11]: weight: 0.103 loss: 0.751
location_date_sensor = 10 [n = 7]: weight: 0.079 loss: 0.980
location_date_sensor = 11 [n = 7]: weight: 0.063 loss: 0.687
location_date_sensor = 12 [n = 16]: weight: 0.034 loss: 0.760
location_date_sensor = 13 [n = 12]: weight: 0.100 loss: 0.855
location_date_sensor = 14 [n = 12]: weight: 0.069 loss: 1.006
location_date_sensor = 15 [n = 13]: weight: 0.074 loss: 0.884
location_date_sensor = 16 [n = 13]: weight: 0.055 loss: 0.719
location_date_sensor = 17 [n = 12]: weight: 0.022 loss: 0.581
44% 399/915 [10:09<13:06, 1.52s/it]objective: 0.157
loss_avg: 0.718
location_date_sensor = 0 [n = 17]: weight: 0.060 loss: 0.748
location_date_sensor = 1 [n = 11]: weight: 0.013 loss: 0.578
location_date_sensor = 2 [n = 15]: weight: 0.030 loss: 0.656
location_date_sensor = 3 [n = 10]: weight: 0.031 loss: 0.547
location_date_sensor = 4 [n = 8]: weight: 0.053 loss: 0.731
location_date_sensor = 5 [n = 10]: weight: 0.023 loss: 0.733
location_date_sensor = 6 [n = 14]: weight: 0.033 loss: 0.694
location_date_sensor = 7 [n = 12]: weight: 0.116 loss: 0.715
location_date_sensor = 8 [n = 9]: weight: 0.045 loss: 0.694
location_date_sensor = 9 [n = 8]: weight: 0.102 loss: 0.742
location_date_sensor = 10 [n = 12]: weight: 0.078 loss: 0.865
location_date_sensor = 11 [n = 17]: weight: 0.063 loss: 0.712
location_date_sensor = 12 [n = 9]: weight: 0.034 loss: 0.691
location_date_sensor = 13 [n = 15]: weight: 0.101 loss: 0.817
location_date_sensor = 14 [n = 6]: weight: 0.069 loss: 0.897
location_date_sensor = 15 [n = 13]: weight: 0.076 loss: 0.826
location_date_sensor = 16 [n = 8]: weight: 0.055 loss: 0.615
location_date_sensor = 17 [n = 6]: weight: 0.021 loss: 0.605
49% 449/915 [11:26<11:53, 1.53s/it]objective: 0.142
loss_avg: 0.715
location_date_sensor = 0 [n = 8]: weight: 0.060 loss: 0.618
location_date_sensor = 1 [n = 7]: weight: 0.012 loss: 0.550
location_date_sensor = 2 [n = 12]: weight: 0.030 loss: 0.656
location_date_sensor = 3 [n = 8]: weight: 0.030 loss: 0.606
location_date_sensor = 4 [n = 12]: weight: 0.053 loss: 0.748
location_date_sensor = 5 [n = 9]: weight: 0.023 loss: 0.606
location_date_sensor = 6 [n = 7]: weight: 0.033 loss: 0.699
location_date_sensor = 7 [n = 2]: weight: 0.113 loss: 0.841
location_date_sensor = 8 [n = 14]: weight: 0.045 loss: 0.713
location_date_sensor = 9 [n = 13]: weight: 0.103 loss: 0.720
location_date_sensor = 10 [n = 10]: weight: 0.079 loss: 0.829
location_date_sensor = 11 [n = 16]: weight: 0.064 loss: 0.682
location_date_sensor = 12 [n = 18]: weight: 0.034 loss: 0.742
location_date_sensor = 13 [n = 9]: weight: 0.101 loss: 0.687
location_date_sensor = 14 [n = 10]: weight: 0.067 loss: 0.868
location_date_sensor = 15 [n = 12]: weight: 0.078 loss: 0.867
location_date_sensor = 16 [n = 13]: weight: 0.056 loss: 0.811
location_date_sensor = 17 [n = 20]: weight: 0.021 loss: 0.640
55% 499/915 [12:42<10:35, 1.53s/it]objective: 0.141
loss_avg: 0.685
location_date_sensor = 0 [n = 10]: weight: 0.060 loss: 0.678
location_date_sensor = 1 [n = 23]: weight: 0.012 loss: 0.530
location_date_sensor = 2 [n = 8]: weight: 0.030 loss: 0.540
location_date_sensor = 3 [n = 11]: weight: 0.029 loss: 0.637
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 0.675
location_date_sensor = 5 [n = 6]: weight: 0.022 loss: 0.657
location_date_sensor = 6 [n = 15]: weight: 0.033 loss: 0.747
location_date_sensor = 7 [n = 10]: weight: 0.110 loss: 0.651
location_date_sensor = 8 [n = 7]: weight: 0.044 loss: 0.611
location_date_sensor = 9 [n = 10]: weight: 0.102 loss: 0.702
location_date_sensor = 10 [n = 11]: weight: 0.079 loss: 0.830
location_date_sensor = 11 [n = 15]: weight: 0.065 loss: 0.677
location_date_sensor = 12 [n = 9]: weight: 0.035 loss: 0.721
location_date_sensor = 13 [n = 14]: weight: 0.102 loss: 0.775
location_date_sensor = 14 [n = 8]: weight: 0.067 loss: 0.857
location_date_sensor = 15 [n = 10]: weight: 0.078 loss: 0.843
location_date_sensor = 16 [n = 11]: weight: 0.057 loss: 0.740
location_date_sensor = 17 [n = 11]: weight: 0.022 loss: 0.567
60% 549/915 [13:58<09:20, 1.53s/it]objective: 0.160
loss_avg: 0.711
location_date_sensor = 0 [n = 3]: weight: 0.057 loss: 0.552
location_date_sensor = 1 [n = 11]: weight: 0.012 loss: 0.481
location_date_sensor = 2 [n = 12]: weight: 0.029 loss: 0.706
location_date_sensor = 3 [n = 10]: weight: 0.030 loss: 0.652
location_date_sensor = 4 [n = 15]: weight: 0.054 loss: 0.682
location_date_sensor = 5 [n = 6]: weight: 0.021 loss: 0.515
location_date_sensor = 6 [n = 10]: weight: 0.032 loss: 0.726
location_date_sensor = 7 [n = 14]: weight: 0.110 loss: 0.720
location_date_sensor = 8 [n = 17]: weight: 0.044 loss: 0.679
location_date_sensor = 9 [n = 13]: weight: 0.103 loss: 0.737
location_date_sensor = 10 [n = 10]: weight: 0.078 loss: 0.855
location_date_sensor = 11 [n = 9]: weight: 0.066 loss: 0.716
location_date_sensor = 12 [n = 11]: weight: 0.035 loss: 0.767
location_date_sensor = 13 [n = 10]: weight: 0.104 loss: 0.798
location_date_sensor = 14 [n = 14]: weight: 0.070 loss: 0.892
location_date_sensor = 15 [n = 10]: weight: 0.079 loss: 0.897
location_date_sensor = 16 [n = 12]: weight: 0.057 loss: 0.630
location_date_sensor = 17 [n = 13]: weight: 0.021 loss: 0.617
65% 599/915 [15:14<08:01, 1.52s/it]objective: 0.145
loss_avg: 0.708
location_date_sensor = 0 [n = 6]: weight: 0.056 loss: 0.596
location_date_sensor = 1 [n = 16]: weight: 0.012 loss: 0.512
location_date_sensor = 2 [n = 18]: weight: 0.030 loss: 0.600
location_date_sensor = 3 [n = 8]: weight: 0.029 loss: 0.630
location_date_sensor = 4 [n = 9]: weight: 0.053 loss: 0.697
location_date_sensor = 5 [n = 15]: weight: 0.021 loss: 0.690
location_date_sensor = 6 [n = 12]: weight: 0.032 loss: 0.734
location_date_sensor = 7 [n = 14]: weight: 0.113 loss: 0.758
location_date_sensor = 8 [n = 6]: weight: 0.044 loss: 0.624
location_date_sensor = 9 [n = 8]: weight: 0.103 loss: 0.694
location_date_sensor = 10 [n = 13]: weight: 0.081 loss: 0.839
location_date_sensor = 11 [n = 16]: weight: 0.066 loss: 0.644
location_date_sensor = 12 [n = 9]: weight: 0.035 loss: 0.740
location_date_sensor = 13 [n = 8]: weight: 0.101 loss: 0.773
location_date_sensor = 14 [n = 10]: weight: 0.071 loss: 0.822
location_date_sensor = 15 [n = 15]: weight: 0.081 loss: 0.889
location_date_sensor = 16 [n = 7]: weight: 0.054 loss: 0.829
location_date_sensor = 17 [n = 10]: weight: 0.021 loss: 0.721
71% 649/915 [16:31<06:46, 1.53s/it]objective: 0.145
loss_avg: 0.705
location_date_sensor = 0 [n = 10]: weight: 0.054 loss: 0.736
location_date_sensor = 1 [n = 11]: weight: 0.012 loss: 0.580
location_date_sensor = 2 [n = 13]: weight: 0.030 loss: 0.614
location_date_sensor = 3 [n = 6]: weight: 0.028 loss: 0.665
location_date_sensor = 4 [n = 11]: weight: 0.053 loss: 0.712
location_date_sensor = 5 [n = 14]: weight: 0.021 loss: 0.667
location_date_sensor = 6 [n = 12]: weight: 0.032 loss: 0.655
location_date_sensor = 7 [n = 13]: weight: 0.115 loss: 0.684
location_date_sensor = 8 [n = 13]: weight: 0.043 loss: 0.704
location_date_sensor = 9 [n = 9]: weight: 0.103 loss: 0.776
location_date_sensor = 10 [n = 11]: weight: 0.082 loss: 0.854
location_date_sensor = 11 [n = 9]: weight: 0.065 loss: 0.646
location_date_sensor = 12 [n = 6]: weight: 0.034 loss: 0.804
location_date_sensor = 13 [n = 11]: weight: 0.100 loss: 0.835
location_date_sensor = 14 [n = 14]: weight: 0.073 loss: 0.853
location_date_sensor = 15 [n = 9]: weight: 0.082 loss: 0.841
location_date_sensor = 16 [n = 10]: weight: 0.054 loss: 0.630
location_date_sensor = 17 [n = 18]: weight: 0.021 loss: 0.564
76% 699/915 [17:47<05:29, 1.52s/it]objective: 0.140
loss_avg: 0.703
location_date_sensor = 0 [n = 9]: weight: 0.053 loss: 0.695
location_date_sensor = 1 [n = 10]: weight: 0.012 loss: 0.536
location_date_sensor = 2 [n = 11]: weight: 0.030 loss: 0.606
location_date_sensor = 3 [n = 17]: weight: 0.028 loss: 0.657
location_date_sensor = 4 [n = 8]: weight: 0.053 loss: 0.645
location_date_sensor = 5 [n = 13]: weight: 0.021 loss: 0.752
location_date_sensor = 6 [n = 9]: weight: 0.032 loss: 0.685
location_date_sensor = 7 [n = 8]: weight: 0.114 loss: 0.629
location_date_sensor = 8 [n = 8]: weight: 0.042 loss: 0.694
location_date_sensor = 9 [n = 8]: weight: 0.102 loss: 0.798
location_date_sensor = 10 [n = 9]: weight: 0.082 loss: 0.859
location_date_sensor = 11 [n = 9]: weight: 0.064 loss: 0.614
location_date_sensor = 12 [n = 12]: weight: 0.034 loss: 0.695
location_date_sensor = 13 [n = 11]: weight: 0.102 loss: 0.784
location_date_sensor = 14 [n = 8]: weight: 0.073 loss: 0.892
location_date_sensor = 15 [n = 17]: weight: 0.085 loss: 0.867
location_date_sensor = 16 [n = 14]: weight: 0.054 loss: 0.671
location_date_sensor = 17 [n = 19]: weight: 0.022 loss: 0.614
82% 749/915 [19:04<04:12, 1.52s/it]objective: 0.143
loss_avg: 0.705
location_date_sensor = 0 [n = 14]: weight: 0.054 loss: 0.800
location_date_sensor = 1 [n = 10]: weight: 0.012 loss: 0.472
location_date_sensor = 2 [n = 9]: weight: 0.029 loss: 0.642
location_date_sensor = 3 [n = 25]: weight: 0.029 loss: 0.668
location_date_sensor = 4 [n = 13]: weight: 0.053 loss: 0.704
location_date_sensor = 5 [n = 16]: weight: 0.022 loss: 0.583
location_date_sensor = 6 [n = 11]: weight: 0.032 loss: 0.713
location_date_sensor = 7 [n = 13]: weight: 0.114 loss: 0.660
location_date_sensor = 8 [n = 6]: weight: 0.042 loss: 0.664
location_date_sensor = 9 [n = 6]: weight: 0.099 loss: 0.720
location_date_sensor = 10 [n = 12]: weight: 0.082 loss: 0.845
location_date_sensor = 11 [n = 13]: weight: 0.063 loss: 0.668
location_date_sensor = 12 [n = 7]: weight: 0.034 loss: 0.690
location_date_sensor = 13 [n = 10]: weight: 0.103 loss: 0.785
location_date_sensor = 14 [n = 13]: weight: 0.074 loss: 0.881
location_date_sensor = 15 [n = 8]: weight: 0.087 loss: 0.844
location_date_sensor = 16 [n = 4]: weight: 0.053 loss: 0.830
location_date_sensor = 17 [n = 10]: weight: 0.022 loss: 0.636
87% 799/915 [20:20<02:57, 1.53s/it]objective: 0.146
loss_avg: 0.695
location_date_sensor = 0 [n = 14]: weight: 0.056 loss: 0.681
location_date_sensor = 1 [n = 7]: weight: 0.011 loss: 0.521
location_date_sensor = 2 [n = 15]: weight: 0.029 loss: 0.637
location_date_sensor = 3 [n = 9]: weight: 0.030 loss: 0.592
location_date_sensor = 4 [n = 5]: weight: 0.052 loss: 0.602
location_date_sensor = 5 [n = 11]: weight: 0.021 loss: 0.437
location_date_sensor = 6 [n = 11]: weight: 0.031 loss: 0.739
location_date_sensor = 7 [n = 12]: weight: 0.113 loss: 0.688
location_date_sensor = 8 [n = 14]: weight: 0.041 loss: 0.683
location_date_sensor = 9 [n = 14]: weight: 0.099 loss: 0.829
location_date_sensor = 10 [n = 12]: weight: 0.084 loss: 0.802
location_date_sensor = 11 [n = 10]: weight: 0.063 loss: 0.616
location_date_sensor = 12 [n = 14]: weight: 0.033 loss: 0.702
location_date_sensor = 13 [n = 9]: weight: 0.103 loss: 0.825
location_date_sensor = 14 [n = 11]: weight: 0.076 loss: 0.843
location_date_sensor = 15 [n = 11]: weight: 0.087 loss: 0.876
location_date_sensor = 16 [n = 10]: weight: 0.052 loss: 0.711
location_date_sensor = 17 [n = 11]: weight: 0.022 loss: 0.596
93% 849/915 [21:37<01:40, 1.53s/it]objective: 0.151
loss_avg: 0.693
location_date_sensor = 0 [n = 14]: weight: 0.056 loss: 0.607
location_date_sensor = 1 [n = 10]: weight: 0.011 loss: 0.551
location_date_sensor = 2 [n = 8]: weight: 0.029 loss: 0.659
location_date_sensor = 3 [n = 11]: weight: 0.030 loss: 0.585
location_date_sensor = 4 [n = 12]: weight: 0.052 loss: 0.641
location_date_sensor = 5 [n = 12]: weight: 0.021 loss: 0.655
location_date_sensor = 6 [n = 13]: weight: 0.032 loss: 0.770
location_date_sensor = 7 [n = 11]: weight: 0.110 loss: 0.712
location_date_sensor = 8 [n = 12]: weight: 0.041 loss: 0.642
location_date_sensor = 9 [n = 9]: weight: 0.097 loss: 0.679
location_date_sensor = 10 [n = 15]: weight: 0.087 loss: 0.812
location_date_sensor = 11 [n = 6]: weight: 0.060 loss: 0.625
location_date_sensor = 12 [n = 9]: weight: 0.033 loss: 0.706
location_date_sensor = 13 [n = 15]: weight: 0.103 loss: 0.795
location_date_sensor = 14 [n = 13]: weight: 0.077 loss: 0.832
location_date_sensor = 15 [n = 10]: weight: 0.088 loss: 0.811
location_date_sensor = 16 [n = 10]: weight: 0.052 loss: 0.681
location_date_sensor = 17 [n = 10]: weight: 0.021 loss: 0.580
98% 899/915 [22:53<00:24, 1.52s/it]objective: 0.148
loss_avg: 0.682
location_date_sensor = 0 [n = 8]: weight: 0.056 loss: 0.623
location_date_sensor = 1 [n = 11]: weight: 0.011 loss: 0.477
location_date_sensor = 2 [n = 12]: weight: 0.029 loss: 0.603
location_date_sensor = 3 [n = 16]: weight: 0.029 loss: 0.555
location_date_sensor = 4 [n = 12]: weight: 0.051 loss: 0.599
location_date_sensor = 5 [n = 9]: weight: 0.021 loss: 0.685
location_date_sensor = 6 [n = 11]: weight: 0.032 loss: 0.726
location_date_sensor = 7 [n = 5]: weight: 0.107 loss: 0.631
location_date_sensor = 8 [n = 10]: weight: 0.041 loss: 0.687
location_date_sensor = 9 [n = 11]: weight: 0.097 loss: 0.700
location_date_sensor = 10 [n = 14]: weight: 0.090 loss: 0.877
location_date_sensor = 11 [n = 15]: weight: 0.060 loss: 0.613
location_date_sensor = 12 [n = 9]: weight: 0.032 loss: 0.756
location_date_sensor = 13 [n = 9]: weight: 0.103 loss: 0.782
location_date_sensor = 14 [n = 10]: weight: 0.080 loss: 0.875
location_date_sensor = 15 [n = 13]: weight: 0.091 loss: 0.829
location_date_sensor = 16 [n = 14]: weight: 0.052 loss: 0.702
location_date_sensor = 17 [n = 11]: weight: 0.021 loss: 0.582
100% 915/915 [23:16<00:00, 1.53s/it]
objective: 0.164
loss_avg: 0.704
location_date_sensor = 0 [n = 1]: weight: 0.054 loss: 0.649
location_date_sensor = 1 [n = 3]: weight: 0.010 loss: 0.561
location_date_sensor = 2 [n = 1]: weight: 0.029 loss: 0.662
location_date_sensor = 3 [n = 4]: weight: 0.029 loss: 0.598
location_date_sensor = 4 [n = 2]: weight: 0.051 loss: 0.646
location_date_sensor = 5 [n = 3]: weight: 0.021 loss: 0.672
location_date_sensor = 6 [n = 4]: weight: 0.032 loss: 0.791
location_date_sensor = 7 [n = 8]: weight: 0.107 loss: 0.645
location_date_sensor = 8 [n = 5]: weight: 0.041 loss: 0.647
location_date_sensor = 9 [n = 5]: weight: 0.097 loss: 0.697
location_date_sensor = 10 [n = 2]: weight: 0.093 loss: 1.099
location_date_sensor = 11 [n = 2]: weight: 0.059 loss: 0.755
location_date_sensor = 12 [n = 4]: weight: 0.032 loss: 0.692
location_date_sensor = 13 [n = 2]: weight: 0.103 loss: 0.696
location_date_sensor = 14 [n = 5]: weight: 0.079 loss: 0.883
location_date_sensor = 15 [n = 1]: weight: 0.091 loss: 0.797
location_date_sensor = 16 [n = 3]: weight: 0.053 loss: 0.685
location_date_sensor = 17 [n = 2]: weight: 0.021 loss: 0.635
Epoch eval:
Average detection_acc: 0.753
location_date_sensor = 0 [n = 186]: detection_acc = 0.747
location_date_sensor = 1 [n = 212]: detection_acc = 0.864
location_date_sensor = 2 [n = 218]: detection_acc = 0.859
location_date_sensor = 3 [n = 217]: detection_acc = 0.804
location_date_sensor = 4 [n = 200]: detection_acc = 0.784
location_date_sensor = 5 [n = 194]: detection_acc = 0.756
location_date_sensor = 6 [n = 186]: detection_acc = 0.752
location_date_sensor = 7 [n = 203]: detection_acc = 0.726
location_date_sensor = 8 [n = 203]: detection_acc = 0.743
location_date_sensor = 9 [n = 194]: detection_acc = 0.623
location_date_sensor = 10 [n = 215]: detection_acc = 0.717
location_date_sensor = 11 [n = 216]: detection_acc = 0.688
location_date_sensor = 12 [n = 189]: detection_acc = 0.745
location_date_sensor = 13 [n = 194]: detection_acc = 0.748
location_date_sensor = 14 [n = 195]: detection_acc = 0.736
location_date_sensor = 15 [n = 203]: detection_acc = 0.775
location_date_sensor = 16 [n = 214]: detection_acc = 0.668
location_date_sensor = 17 [n = 218]: detection_acc = 0.814
Worst-group detection_acc: 0.623
Validation:
100% 348/348 [03:19<00:00, 1.74it/s]
objective: 0.000
loss_avg: 1.000
location_date_sensor = 18 [n = 12]: weight: 0.000 loss: 1.000
location_date_sensor = 19 [n = 49]: weight: 0.000 loss: 1.000
location_date_sensor = 20 [n = 254]: weight: 0.000 loss: 1.000
location_date_sensor = 21 [n = 216]: weight: 0.000 loss: 1.000
location_date_sensor = 22 [n = 89]: weight: 0.000 loss: 1.000
location_date_sensor = 23 [n = 11]: weight: 0.000 loss: 1.000
location_date_sensor = 24 [n = 4]: weight: 0.000 loss: 1.000
location_date_sensor = 25 [n = 7]: weight: 0.000 loss: 1.000
location_date_sensor = 26 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 27 [n = 17]: weight: 0.000 loss: 1.000
location_date_sensor = 28 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 29 [n = 8]: weight: 0.000 loss: 1.000
location_date_sensor = 30 [n = 43]: weight: 0.000 loss: 1.000
location_date_sensor = 31 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 32 [n = 51]: weight: 0.000 loss: 1.000
location_date_sensor = 33 [n = 50]: weight: 0.000 loss: 1.000
location_date_sensor = 34 [n = 28]: weight: 0.000 loss: 1.000
location_date_sensor = 35 [n = 75]: weight: 0.000 loss: 1.000
location_date_sensor = 36 [n = 56]: weight: 0.000 loss: 1.000
location_date_sensor = 37 [n = 34]: weight: 0.000 loss: 1.000
location_date_sensor = 38 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 39 [n = 13]: weight: 0.000 loss: 1.000
location_date_sensor = 40 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 41 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 42 [n = 57]: weight: 0.000 loss: 1.000
location_date_sensor = 43 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 44 [n = 33]: weight: 0.000 loss: 1.000
location_date_sensor = 45 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 46 [n = 30]: weight: 0.000 loss: 1.000
Epoch eval:
Average detection_acc: 0.014
location_date_sensor = 18 [n = 12]: detection_acc = 0.000
location_date_sensor = 19 [n = 49]: detection_acc = 0.020
location_date_sensor = 20 [n = 254]: detection_acc = 0.000
location_date_sensor = 21 [n = 216]: detection_acc = 0.000
location_date_sensor = 22 [n = 89]: detection_acc = 0.000
location_date_sensor = 23 [n = 11]: detection_acc = 0.000
location_date_sensor = 24 [n = 4]: detection_acc = 0.500
location_date_sensor = 25 [n = 7]: detection_acc = 0.143
location_date_sensor = 26 [n = 14]: detection_acc = 0.071
location_date_sensor = 27 [n = 17]: detection_acc = 0.059
location_date_sensor = 28 [n = 14]: detection_acc = 0.000
location_date_sensor = 29 [n = 8]: detection_acc = 0.000
location_date_sensor = 30 [n = 43]: detection_acc = 0.000
location_date_sensor = 31 [n = 55]: detection_acc = 0.000
location_date_sensor = 32 [n = 51]: detection_acc = 0.000
location_date_sensor = 33 [n = 50]: detection_acc = 0.000
location_date_sensor = 34 [n = 28]: detection_acc = 0.000
location_date_sensor = 35 [n = 75]: detection_acc = 0.040
location_date_sensor = 36 [n = 56]: detection_acc = 0.054
location_date_sensor = 37 [n = 34]: detection_acc = 0.029
location_date_sensor = 38 [n = 55]: detection_acc = 0.000
location_date_sensor = 39 [n = 13]: detection_acc = 0.000
location_date_sensor = 40 [n = 19]: detection_acc = 0.000
location_date_sensor = 41 [n = 19]: detection_acc = 0.000
location_date_sensor = 42 [n = 57]: detection_acc = 0.000
location_date_sensor = 43 [n = 39]: detection_acc = 0.000
location_date_sensor = 44 [n = 33]: detection_acc = 0.030
location_date_sensor = 45 [n = 39]: detection_acc = 0.026
location_date_sensor = 46 [n = 30]: detection_acc = 0.133
Worst-group detection_acc: 0.000
Validation detection_acc_avg: 0.014
Epoch 2 has the best validation performance so far.
100% 365/365 [03:29<00:00, 1.74it/s]
Epoch [3]:
Train:
5% 49/915 [01:15<22:04, 1.53s/it]objective: 0.146
loss_avg: 0.694
location_date_sensor = 0 [n = 14]: weight: 0.054 loss: 0.587
location_date_sensor = 1 [n = 5]: weight: 0.010 loss: 0.587
location_date_sensor = 2 [n = 13]: weight: 0.028 loss: 0.587
location_date_sensor = 3 [n = 7]: weight: 0.029 loss: 0.577
location_date_sensor = 4 [n = 14]: weight: 0.051 loss: 0.679
location_date_sensor = 5 [n = 11]: weight: 0.021 loss: 0.887
location_date_sensor = 6 [n = 17]: weight: 0.032 loss: 0.692
location_date_sensor = 7 [n = 7]: weight: 0.106 loss: 0.617
location_date_sensor = 8 [n = 11]: weight: 0.041 loss: 0.679
location_date_sensor = 9 [n = 9]: weight: 0.096 loss: 0.681
location_date_sensor = 10 [n = 14]: weight: 0.093 loss: 0.814
location_date_sensor = 11 [n = 4]: weight: 0.058 loss: 0.596
location_date_sensor = 12 [n = 12]: weight: 0.032 loss: 0.708
location_date_sensor = 13 [n = 14]: weight: 0.104 loss: 0.718
location_date_sensor = 14 [n = 16]: weight: 0.082 loss: 0.813
location_date_sensor = 15 [n = 7]: weight: 0.091 loss: 0.910
location_date_sensor = 16 [n = 9]: weight: 0.053 loss: 0.605
location_date_sensor = 17 [n = 16]: weight: 0.021 loss: 0.620
11% 99/915 [02:31<20:52, 1.54s/it]objective: 0.147
loss_avg: 0.704
location_date_sensor = 0 [n = 8]: weight: 0.054 loss: 0.628
location_date_sensor = 1 [n = 11]: weight: 0.010 loss: 0.537
location_date_sensor = 2 [n = 11]: weight: 0.028 loss: 0.649
location_date_sensor = 3 [n = 20]: weight: 0.029 loss: 0.631
location_date_sensor = 4 [n = 13]: weight: 0.052 loss: 0.631
location_date_sensor = 5 [n = 11]: weight: 0.021 loss: 0.822
location_date_sensor = 6 [n = 11]: weight: 0.033 loss: 0.733
location_date_sensor = 7 [n = 11]: weight: 0.105 loss: 0.728
location_date_sensor = 8 [n = 8]: weight: 0.041 loss: 0.635
location_date_sensor = 9 [n = 12]: weight: 0.095 loss: 0.727
location_date_sensor = 10 [n = 10]: weight: 0.095 loss: 0.784
location_date_sensor = 11 [n = 9]: weight: 0.056 loss: 0.612
location_date_sensor = 12 [n = 18]: weight: 0.033 loss: 0.700
location_date_sensor = 13 [n = 8]: weight: 0.103 loss: 0.707
location_date_sensor = 14 [n = 9]: weight: 0.085 loss: 0.949
location_date_sensor = 15 [n = 12]: weight: 0.091 loss: 0.900
location_date_sensor = 16 [n = 9]: weight: 0.051 loss: 0.692
location_date_sensor = 17 [n = 9]: weight: 0.021 loss: 0.638
16% 149/915 [03:48<19:29, 1.53s/it]objective: 0.147
loss_avg: 0.698
location_date_sensor = 0 [n = 16]: weight: 0.053 loss: 0.621
location_date_sensor = 1 [n = 13]: weight: 0.010 loss: 0.542
location_date_sensor = 2 [n = 11]: weight: 0.028 loss: 0.620
location_date_sensor = 3 [n = 13]: weight: 0.029 loss: 0.575
location_date_sensor = 4 [n = 6]: weight: 0.051 loss: 0.637
location_date_sensor = 5 [n = 9]: weight: 0.021 loss: 0.678
location_date_sensor = 6 [n = 12]: weight: 0.033 loss: 0.779
location_date_sensor = 7 [n = 6]: weight: 0.105 loss: 0.693
location_date_sensor = 8 [n = 11]: weight: 0.040 loss: 0.669
location_date_sensor = 9 [n = 9]: weight: 0.094 loss: 0.686
location_date_sensor = 10 [n = 16]: weight: 0.096 loss: 0.876
location_date_sensor = 11 [n = 17]: weight: 0.056 loss: 0.612
location_date_sensor = 12 [n = 12]: weight: 0.034 loss: 0.721
location_date_sensor = 13 [n = 13]: weight: 0.105 loss: 0.753
location_date_sensor = 14 [n = 12]: weight: 0.086 loss: 1.014
location_date_sensor = 15 [n = 8]: weight: 0.091 loss: 0.793
location_date_sensor = 16 [n = 6]: weight: 0.051 loss: 0.630
location_date_sensor = 17 [n = 10]: weight: 0.021 loss: 0.627
22% 199/915 [05:04<18:10, 1.52s/it]objective: 0.148
loss_avg: 0.712
location_date_sensor = 0 [n = 5]: weight: 0.053 loss: 0.708
location_date_sensor = 1 [n = 14]: weight: 0.010 loss: 0.504
location_date_sensor = 2 [n = 10]: weight: 0.028 loss: 0.606
location_date_sensor = 3 [n = 12]: weight: 0.029 loss: 0.628
location_date_sensor = 4 [n = 19]: weight: 0.050 loss: 0.623
location_date_sensor = 5 [n = 8]: weight: 0.020 loss: 0.650
location_date_sensor = 6 [n = 13]: weight: 0.033 loss: 0.730
location_date_sensor = 7 [n = 10]: weight: 0.103 loss: 0.779
location_date_sensor = 8 [n = 4]: weight: 0.039 loss: 0.662
location_date_sensor = 9 [n = 9]: weight: 0.093 loss: 0.696
location_date_sensor = 10 [n = 14]: weight: 0.099 loss: 0.920
location_date_sensor = 11 [n = 6]: weight: 0.055 loss: 0.680
location_date_sensor = 12 [n = 14]: weight: 0.034 loss: 0.742
location_date_sensor = 13 [n = 10]: weight: 0.106 loss: 0.816
location_date_sensor = 14 [n = 10]: weight: 0.087 loss: 0.931
location_date_sensor = 15 [n = 10]: weight: 0.092 loss: 0.888
location_date_sensor = 16 [n = 18]: weight: 0.051 loss: 0.705
location_date_sensor = 17 [n = 14]: weight: 0.021 loss: 0.609
27% 249/915 [06:21<17:00, 1.53s/it]objective: 0.158
loss_avg: 0.683
location_date_sensor = 0 [n = 15]: weight: 0.051 loss: 0.611
location_date_sensor = 1 [n = 8]: weight: 0.009 loss: 0.520
location_date_sensor = 2 [n = 16]: weight: 0.028 loss: 0.568
location_date_sensor = 3 [n = 11]: weight: 0.029 loss: 0.569
location_date_sensor = 4 [n = 6]: weight: 0.051 loss: 0.623
location_date_sensor = 5 [n = 6]: weight: 0.020 loss: 0.731
location_date_sensor = 6 [n = 16]: weight: 0.034 loss: 0.722
location_date_sensor = 7 [n = 13]: weight: 0.103 loss: 0.635
location_date_sensor = 8 [n = 9]: weight: 0.038 loss: 0.674
location_date_sensor = 9 [n = 15]: weight: 0.094 loss: 0.701
location_date_sensor = 10 [n = 13]: weight: 0.101 loss: 0.844
location_date_sensor = 11 [n = 6]: weight: 0.053 loss: 0.626
location_date_sensor = 12 [n = 10]: weight: 0.034 loss: 0.703
location_date_sensor = 13 [n = 14]: weight: 0.107 loss: 0.739
location_date_sensor = 14 [n = 13]: weight: 0.089 loss: 0.868
location_date_sensor = 15 [n = 9]: weight: 0.091 loss: 0.799
location_date_sensor = 16 [n = 9]: weight: 0.051 loss: 0.652
location_date_sensor = 17 [n = 11]: weight: 0.020 loss: 0.627
33% 299/915 [07:37<15:39, 1.53s/it]objective: 0.148
loss_avg: 0.677
location_date_sensor = 0 [n = 19]: weight: 0.052 loss: 0.628
location_date_sensor = 1 [n = 7]: weight: 0.009 loss: 0.504
location_date_sensor = 2 [n = 15]: weight: 0.028 loss: 0.575
location_date_sensor = 3 [n = 10]: weight: 0.028 loss: 0.591
location_date_sensor = 4 [n = 9]: weight: 0.049 loss: 0.613
location_date_sensor = 5 [n = 9]: weight: 0.019 loss: 0.561
location_date_sensor = 6 [n = 23]: weight: 0.035 loss: 0.727
location_date_sensor = 7 [n = 12]: weight: 0.103 loss: 0.661
location_date_sensor = 8 [n = 7]: weight: 0.037 loss: 0.637
location_date_sensor = 9 [n = 10]: weight: 0.095 loss: 0.658
location_date_sensor = 10 [n = 6]: weight: 0.101 loss: 0.878
location_date_sensor = 11 [n = 4]: weight: 0.051 loss: 0.588
location_date_sensor = 12 [n = 8]: weight: 0.034 loss: 0.746
location_date_sensor = 13 [n = 10]: weight: 0.107 loss: 0.744
location_date_sensor = 14 [n = 14]: weight: 0.094 loss: 0.907
location_date_sensor = 15 [n = 17]: weight: 0.092 loss: 0.811
location_date_sensor = 16 [n = 12]: weight: 0.050 loss: 0.618
location_date_sensor = 17 [n = 8]: weight: 0.020 loss: 0.535
38% 349/915 [08:54<14:24, 1.53s/it]objective: 0.150
loss_avg: 0.677
location_date_sensor = 0 [n = 6]: weight: 0.053 loss: 0.786
location_date_sensor = 1 [n = 9]: weight: 0.009 loss: 0.542
location_date_sensor = 2 [n = 10]: weight: 0.027 loss: 0.641
location_date_sensor = 3 [n = 3]: weight: 0.027 loss: 0.652
location_date_sensor = 4 [n = 11]: weight: 0.048 loss: 0.622
location_date_sensor = 5 [n = 7]: weight: 0.018 loss: 0.633
location_date_sensor = 6 [n = 13]: weight: 0.036 loss: 0.672
location_date_sensor = 7 [n = 13]: weight: 0.103 loss: 0.639
location_date_sensor = 8 [n = 16]: weight: 0.036 loss: 0.597
location_date_sensor = 9 [n = 16]: weight: 0.094 loss: 0.704
location_date_sensor = 10 [n = 11]: weight: 0.101 loss: 0.876
location_date_sensor = 11 [n = 15]: weight: 0.050 loss: 0.592
location_date_sensor = 12 [n = 15]: weight: 0.033 loss: 0.698
location_date_sensor = 13 [n = 14]: weight: 0.109 loss: 0.723
location_date_sensor = 14 [n = 11]: weight: 0.096 loss: 0.786
location_date_sensor = 15 [n = 8]: weight: 0.093 loss: 0.870
location_date_sensor = 16 [n = 10]: weight: 0.049 loss: 0.624
location_date_sensor = 17 [n = 12]: weight: 0.020 loss: 0.617
44% 399/915 [10:10<13:06, 1.52s/it]objective: 0.148
loss_avg: 0.662
location_date_sensor = 0 [n = 10]: weight: 0.051 loss: 0.559
location_date_sensor = 1 [n = 7]: weight: 0.008 loss: 0.481
location_date_sensor = 2 [n = 15]: weight: 0.027 loss: 0.598
location_date_sensor = 3 [n = 12]: weight: 0.027 loss: 0.606
location_date_sensor = 4 [n = 11]: weight: 0.048 loss: 0.605
location_date_sensor = 5 [n = 10]: weight: 0.018 loss: 0.565
location_date_sensor = 6 [n = 9]: weight: 0.036 loss: 0.619
location_date_sensor = 7 [n = 14]: weight: 0.102 loss: 0.623
location_date_sensor = 8 [n = 5]: weight: 0.036 loss: 0.668
location_date_sensor = 9 [n = 14]: weight: 0.095 loss: 0.714
location_date_sensor = 10 [n = 12]: weight: 0.102 loss: 0.844
location_date_sensor = 11 [n = 10]: weight: 0.048 loss: 0.624
location_date_sensor = 12 [n = 13]: weight: 0.034 loss: 0.716
location_date_sensor = 13 [n = 12]: weight: 0.111 loss: 0.645
location_date_sensor = 14 [n = 6]: weight: 0.094 loss: 0.874
location_date_sensor = 15 [n = 18]: weight: 0.097 loss: 0.891
location_date_sensor = 16 [n = 9]: weight: 0.049 loss: 0.574
location_date_sensor = 17 [n = 13]: weight: 0.019 loss: 0.563
49% 449/915 [11:27<11:51, 1.53s/it]objective: 0.137
loss_avg: 0.656
location_date_sensor = 0 [n = 5]: weight: 0.049 loss: 0.581
location_date_sensor = 1 [n = 13]: weight: 0.008 loss: 0.509
location_date_sensor = 2 [n = 11]: weight: 0.027 loss: 0.568
location_date_sensor = 3 [n = 20]: weight: 0.027 loss: 0.514
location_date_sensor = 4 [n = 4]: weight: 0.046 loss: 0.648
location_date_sensor = 5 [n = 15]: weight: 0.018 loss: 0.539
location_date_sensor = 6 [n = 17]: weight: 0.035 loss: 0.642
location_date_sensor = 7 [n = 7]: weight: 0.101 loss: 0.589
location_date_sensor = 8 [n = 11]: weight: 0.036 loss: 0.638
location_date_sensor = 9 [n = 14]: weight: 0.096 loss: 0.709
location_date_sensor = 10 [n = 15]: weight: 0.105 loss: 0.820
location_date_sensor = 11 [n = 7]: weight: 0.048 loss: 0.647
location_date_sensor = 12 [n = 11]: weight: 0.034 loss: 0.761
location_date_sensor = 13 [n = 12]: weight: 0.112 loss: 0.775
location_date_sensor = 14 [n = 6]: weight: 0.093 loss: 0.761
location_date_sensor = 15 [n = 11]: weight: 0.101 loss: 0.901
location_date_sensor = 16 [n = 7]: weight: 0.047 loss: 0.804
location_date_sensor = 17 [n = 14]: weight: 0.019 loss: 0.568
55% 499/915 [12:43<10:37, 1.53s/it]objective: 0.152
loss_avg: 0.663
location_date_sensor = 0 [n = 19]: weight: 0.049 loss: 0.581
location_date_sensor = 1 [n = 13]: weight: 0.008 loss: 0.499
location_date_sensor = 2 [n = 7]: weight: 0.027 loss: 0.523
location_date_sensor = 3 [n = 12]: weight: 0.027 loss: 0.622
location_date_sensor = 4 [n = 7]: weight: 0.044 loss: 0.620
location_date_sensor = 5 [n = 10]: weight: 0.017 loss: 0.502
location_date_sensor = 6 [n = 9]: weight: 0.036 loss: 0.721
location_date_sensor = 7 [n = 11]: weight: 0.099 loss: 0.662
location_date_sensor = 8 [n = 6]: weight: 0.035 loss: 0.667
location_date_sensor = 9 [n = 12]: weight: 0.096 loss: 0.653
location_date_sensor = 10 [n = 9]: weight: 0.109 loss: 0.813
location_date_sensor = 11 [n = 12]: weight: 0.047 loss: 0.582
location_date_sensor = 12 [n = 14]: weight: 0.034 loss: 0.695
location_date_sensor = 13 [n = 11]: weight: 0.111 loss: 0.762
location_date_sensor = 14 [n = 7]: weight: 0.092 loss: 0.928
location_date_sensor = 15 [n = 18]: weight: 0.105 loss: 0.836
location_date_sensor = 16 [n = 12]: weight: 0.048 loss: 0.696
location_date_sensor = 17 [n = 11]: weight: 0.019 loss: 0.611
60% 549/915 [14:00<09:18, 1.53s/it]objective: 0.135
loss_avg: 0.659
location_date_sensor = 0 [n = 11]: weight: 0.049 loss: 0.542
location_date_sensor = 1 [n = 13]: weight: 0.008 loss: 0.544
location_date_sensor = 2 [n = 12]: weight: 0.026 loss: 0.591
location_date_sensor = 3 [n = 14]: weight: 0.027 loss: 0.581
location_date_sensor = 4 [n = 13]: weight: 0.045 loss: 0.622
location_date_sensor = 5 [n = 9]: weight: 0.018 loss: 0.713
location_date_sensor = 6 [n = 13]: weight: 0.036 loss: 0.673
location_date_sensor = 7 [n = 12]: weight: 0.098 loss: 0.612
location_date_sensor = 8 [n = 7]: weight: 0.034 loss: 0.659
location_date_sensor = 9 [n = 12]: weight: 0.096 loss: 0.640
location_date_sensor = 10 [n = 7]: weight: 0.108 loss: 0.768
location_date_sensor = 11 [n = 9]: weight: 0.046 loss: 0.567
location_date_sensor = 12 [n = 10]: weight: 0.035 loss: 0.680
location_date_sensor = 13 [n = 6]: weight: 0.108 loss: 0.886
location_date_sensor = 14 [n = 6]: weight: 0.090 loss: 0.887
location_date_sensor = 15 [n = 14]: weight: 0.113 loss: 0.816
location_date_sensor = 16 [n = 18]: weight: 0.048 loss: 0.717
location_date_sensor = 17 [n = 14]: weight: 0.019 loss: 0.601
65% 599/915 [15:16<08:04, 1.53s/it]objective: 0.152
loss_avg: 0.661
location_date_sensor = 0 [n = 6]: weight: 0.049 loss: 0.572
location_date_sensor = 1 [n = 17]: weight: 0.008 loss: 0.498
location_date_sensor = 2 [n = 10]: weight: 0.026 loss: 0.641
location_date_sensor = 3 [n = 10]: weight: 0.027 loss: 0.553
location_date_sensor = 4 [n = 14]: weight: 0.044 loss: 0.610
location_date_sensor = 5 [n = 11]: weight: 0.017 loss: 0.603
location_date_sensor = 6 [n = 10]: weight: 0.036 loss: 0.653
location_date_sensor = 7 [n = 10]: weight: 0.096 loss: 0.658
location_date_sensor = 8 [n = 8]: weight: 0.033 loss: 0.660
location_date_sensor = 9 [n = 7]: weight: 0.095 loss: 0.595
location_date_sensor = 10 [n = 10]: weight: 0.107 loss: 0.852
location_date_sensor = 11 [n = 8]: weight: 0.045 loss: 0.572
location_date_sensor = 12 [n = 9]: weight: 0.034 loss: 0.634
location_date_sensor = 13 [n = 14]: weight: 0.109 loss: 0.666
location_date_sensor = 14 [n = 17]: weight: 0.091 loss: 0.822
location_date_sensor = 15 [n = 18]: weight: 0.118 loss: 0.828
location_date_sensor = 16 [n = 11]: weight: 0.049 loss: 0.684
location_date_sensor = 17 [n = 10]: weight: 0.019 loss: 0.609
71% 649/915 [16:33<06:44, 1.52s/it]objective: 0.142
loss_avg: 0.673
location_date_sensor = 0 [n = 14]: weight: 0.048 loss: 0.650
location_date_sensor = 1 [n = 7]: weight: 0.008 loss: 0.489
location_date_sensor = 2 [n = 16]: weight: 0.026 loss: 0.627
location_date_sensor = 3 [n = 5]: weight: 0.026 loss: 0.526
location_date_sensor = 4 [n = 5]: weight: 0.043 loss: 0.594
location_date_sensor = 5 [n = 12]: weight: 0.017 loss: 0.715
location_date_sensor = 6 [n = 14]: weight: 0.036 loss: 0.691
location_date_sensor = 7 [n = 13]: weight: 0.096 loss: 0.591
location_date_sensor = 8 [n = 13]: weight: 0.033 loss: 0.623
location_date_sensor = 9 [n = 9]: weight: 0.093 loss: 0.663
location_date_sensor = 10 [n = 11]: weight: 0.107 loss: 0.838
location_date_sensor = 11 [n = 10]: weight: 0.044 loss: 0.603
location_date_sensor = 12 [n = 13]: weight: 0.035 loss: 0.762
location_date_sensor = 13 [n = 16]: weight: 0.110 loss: 0.655
location_date_sensor = 14 [n = 7]: weight: 0.092 loss: 0.891
location_date_sensor = 15 [n = 15]: weight: 0.123 loss: 0.812
location_date_sensor = 16 [n = 9]: weight: 0.048 loss: 0.602
location_date_sensor = 17 [n = 11]: weight: 0.019 loss: 0.636
76% 699/915 [17:49<05:30, 1.53s/it]objective: 0.151
loss_avg: 0.664
location_date_sensor = 0 [n = 10]: weight: 0.048 loss: 0.589
location_date_sensor = 1 [n = 14]: weight: 0.007 loss: 0.516
location_date_sensor = 2 [n = 8]: weight: 0.026 loss: 0.575
location_date_sensor = 3 [n = 15]: weight: 0.025 loss: 0.502
location_date_sensor = 4 [n = 11]: weight: 0.042 loss: 0.594
location_date_sensor = 5 [n = 10]: weight: 0.017 loss: 0.654
location_date_sensor = 6 [n = 5]: weight: 0.037 loss: 0.710
location_date_sensor = 7 [n = 8]: weight: 0.094 loss: 0.629
location_date_sensor = 8 [n = 13]: weight: 0.033 loss: 0.653
location_date_sensor = 9 [n = 4]: weight: 0.088 loss: 0.674
location_date_sensor = 10 [n = 15]: weight: 0.110 loss: 0.848
location_date_sensor = 11 [n = 10]: weight: 0.043 loss: 0.554
location_date_sensor = 12 [n = 11]: weight: 0.035 loss: 0.701
location_date_sensor = 13 [n = 9]: weight: 0.109 loss: 0.695
location_date_sensor = 14 [n = 17]: weight: 0.093 loss: 0.833
location_date_sensor = 15 [n = 14]: weight: 0.128 loss: 0.866
location_date_sensor = 16 [n = 14]: weight: 0.048 loss: 0.634
location_date_sensor = 17 [n = 12]: weight: 0.019 loss: 0.602
82% 749/915 [19:06<04:14, 1.53s/it]objective: 0.131
loss_avg: 0.635
location_date_sensor = 0 [n = 15]: weight: 0.048 loss: 0.589
location_date_sensor = 1 [n = 15]: weight: 0.007 loss: 0.467
location_date_sensor = 2 [n = 11]: weight: 0.025 loss: 0.570
location_date_sensor = 3 [n = 17]: weight: 0.025 loss: 0.543
location_date_sensor = 4 [n = 3]: weight: 0.040 loss: 0.556
location_date_sensor = 5 [n = 13]: weight: 0.017 loss: 0.667
location_date_sensor = 6 [n = 10]: weight: 0.036 loss: 0.683
location_date_sensor = 7 [n = 9]: weight: 0.092 loss: 0.559
location_date_sensor = 8 [n = 10]: weight: 0.033 loss: 0.616
location_date_sensor = 9 [n = 13]: weight: 0.088 loss: 0.665
location_date_sensor = 10 [n = 10]: weight: 0.112 loss: 0.768
location_date_sensor = 11 [n = 12]: weight: 0.043 loss: 0.561
location_date_sensor = 12 [n = 17]: weight: 0.035 loss: 0.705
location_date_sensor = 13 [n = 8]: weight: 0.107 loss: 0.645
location_date_sensor = 14 [n = 8]: weight: 0.095 loss: 0.807
location_date_sensor = 15 [n = 14]: weight: 0.131 loss: 0.816
location_date_sensor = 16 [n = 8]: weight: 0.048 loss: 0.662
location_date_sensor = 17 [n = 7]: weight: 0.019 loss: 0.562
87% 799/915 [20:22<02:57, 1.53s/it]objective: 0.160
loss_avg: 0.660
location_date_sensor = 0 [n = 10]: weight: 0.047 loss: 0.524
location_date_sensor = 1 [n = 13]: weight: 0.007 loss: 0.589
location_date_sensor = 2 [n = 11]: weight: 0.025 loss: 0.552
location_date_sensor = 3 [n = 6]: weight: 0.026 loss: 0.552
location_date_sensor = 4 [n = 10]: weight: 0.039 loss: 0.652
location_date_sensor = 5 [n = 9]: weight: 0.017 loss: 0.603
location_date_sensor = 6 [n = 5]: weight: 0.035 loss: 0.750
location_date_sensor = 7 [n = 12]: weight: 0.090 loss: 0.647
location_date_sensor = 8 [n = 10]: weight: 0.033 loss: 0.619
location_date_sensor = 9 [n = 8]: weight: 0.088 loss: 0.616
location_date_sensor = 10 [n = 11]: weight: 0.112 loss: 0.824
location_date_sensor = 11 [n = 14]: weight: 0.043 loss: 0.570
location_date_sensor = 12 [n = 7]: weight: 0.035 loss: 0.820
location_date_sensor = 13 [n = 20]: weight: 0.109 loss: 0.711
location_date_sensor = 14 [n = 10]: weight: 0.097 loss: 0.857
location_date_sensor = 15 [n = 13]: weight: 0.135 loss: 0.861
location_date_sensor = 16 [n = 10]: weight: 0.047 loss: 0.607
location_date_sensor = 17 [n = 21]: weight: 0.019 loss: 0.582
93% 849/915 [21:39<01:40, 1.53s/it]objective: 0.137
loss_avg: 0.634
location_date_sensor = 0 [n = 12]: weight: 0.047 loss: 0.537
location_date_sensor = 1 [n = 11]: weight: 0.007 loss: 0.531
location_date_sensor = 2 [n = 13]: weight: 0.025 loss: 0.577
location_date_sensor = 3 [n = 15]: weight: 0.025 loss: 0.525
location_date_sensor = 4 [n = 10]: weight: 0.038 loss: 0.572
location_date_sensor = 5 [n = 13]: weight: 0.017 loss: 0.689
location_date_sensor = 6 [n = 7]: weight: 0.034 loss: 0.670
location_date_sensor = 7 [n = 10]: weight: 0.089 loss: 0.571
location_date_sensor = 8 [n = 15]: weight: 0.032 loss: 0.615
location_date_sensor = 9 [n = 12]: weight: 0.086 loss: 0.709
location_date_sensor = 10 [n = 11]: weight: 0.113 loss: 0.748
location_date_sensor = 11 [n = 15]: weight: 0.042 loss: 0.522
location_date_sensor = 12 [n = 11]: weight: 0.036 loss: 0.696
location_date_sensor = 13 [n = 6]: weight: 0.111 loss: 0.632
location_date_sensor = 14 [n = 11]: weight: 0.098 loss: 0.798
location_date_sensor = 15 [n = 14]: weight: 0.139 loss: 0.824
location_date_sensor = 16 [n = 5]: weight: 0.045 loss: 0.686
location_date_sensor = 17 [n = 9]: weight: 0.018 loss: 0.575
98% 899/915 [22:55<00:24, 1.53s/it]objective: 0.163
loss_avg: 0.675
location_date_sensor = 0 [n = 8]: weight: 0.045 loss: 0.546
location_date_sensor = 1 [n = 13]: weight: 0.007 loss: 0.530
location_date_sensor = 2 [n = 13]: weight: 0.025 loss: 0.593
location_date_sensor = 3 [n = 9]: weight: 0.024 loss: 0.535
location_date_sensor = 4 [n = 10]: weight: 0.038 loss: 0.618
location_date_sensor = 5 [n = 11]: weight: 0.017 loss: 0.639
location_date_sensor = 6 [n = 15]: weight: 0.033 loss: 0.710
location_date_sensor = 7 [n = 11]: weight: 0.087 loss: 0.620
location_date_sensor = 8 [n = 9]: weight: 0.032 loss: 0.643
location_date_sensor = 9 [n = 7]: weight: 0.083 loss: 0.700
location_date_sensor = 10 [n = 16]: weight: 0.116 loss: 0.790
location_date_sensor = 11 [n = 11]: weight: 0.042 loss: 0.578
location_date_sensor = 12 [n = 14]: weight: 0.036 loss: 0.664
location_date_sensor = 13 [n = 14]: weight: 0.111 loss: 0.757
location_date_sensor = 14 [n = 14]: weight: 0.099 loss: 0.820
location_date_sensor = 15 [n = 12]: weight: 0.144 loss: 0.882
location_date_sensor = 16 [n = 6]: weight: 0.043 loss: 0.679
location_date_sensor = 17 [n = 7]: weight: 0.018 loss: 0.676
100% 915/915 [23:19<00:00, 1.53s/it]
objective: 0.139
loss_avg: 0.667
location_date_sensor = 0 [n = 3]: weight: 0.045 loss: 0.672
location_date_sensor = 1 [n = 2]: weight: 0.007 loss: 0.653
location_date_sensor = 2 [n = 6]: weight: 0.024 loss: 0.551
location_date_sensor = 3 [n = 2]: weight: 0.024 loss: 0.618
location_date_sensor = 4 [n = 1]: weight: 0.037 loss: 0.758
location_date_sensor = 6 [n = 1]: weight: 0.034 loss: 0.675
location_date_sensor = 7 [n = 3]: weight: 0.088 loss: 0.658
location_date_sensor = 8 [n = 4]: weight: 0.032 loss: 0.608
location_date_sensor = 9 [n = 4]: weight: 0.083 loss: 0.681
location_date_sensor = 10 [n = 1]: weight: 0.117 loss: 0.775
location_date_sensor = 11 [n = 3]: weight: 0.042 loss: 0.551
location_date_sensor = 12 [n = 7]: weight: 0.036 loss: 0.720
location_date_sensor = 13 [n = 3]: weight: 0.111 loss: 0.686
location_date_sensor = 14 [n = 4]: weight: 0.101 loss: 0.834
location_date_sensor = 15 [n = 3]: weight: 0.144 loss: 0.877
location_date_sensor = 16 [n = 4]: weight: 0.043 loss: 0.623
location_date_sensor = 17 [n = 6]: weight: 0.018 loss: 0.597
Epoch eval:
Average detection_acc: 0.782
location_date_sensor = 0 [n = 206]: detection_acc = 0.823
location_date_sensor = 1 [n = 205]: detection_acc = 0.874
location_date_sensor = 2 [n = 219]: detection_acc = 0.880
location_date_sensor = 3 [n = 213]: detection_acc = 0.827
location_date_sensor = 4 [n = 177]: detection_acc = 0.815
location_date_sensor = 5 [n = 184]: detection_acc = 0.792
location_date_sensor = 6 [n = 220]: detection_acc = 0.767
location_date_sensor = 7 [n = 192]: detection_acc = 0.773
location_date_sensor = 8 [n = 177]: detection_acc = 0.770
location_date_sensor = 9 [n = 196]: detection_acc = 0.641
location_date_sensor = 10 [n = 212]: detection_acc = 0.741
location_date_sensor = 11 [n = 182]: detection_acc = 0.733
location_date_sensor = 12 [n = 226]: detection_acc = 0.745
location_date_sensor = 13 [n = 214]: detection_acc = 0.808
location_date_sensor = 14 [n = 198]: detection_acc = 0.752
location_date_sensor = 15 [n = 235]: detection_acc = 0.772
location_date_sensor = 16 [n = 186]: detection_acc = 0.722
location_date_sensor = 17 [n = 215]: detection_acc = 0.820
Worst-group detection_acc: 0.641
Validation:
100% 348/348 [03:19<00:00, 1.74it/s]
objective: 0.000
loss_avg: 1.000
location_date_sensor = 18 [n = 12]: weight: 0.000 loss: 1.000
location_date_sensor = 19 [n = 49]: weight: 0.000 loss: 1.000
location_date_sensor = 20 [n = 254]: weight: 0.000 loss: 1.000
location_date_sensor = 21 [n = 216]: weight: 0.000 loss: 1.000
location_date_sensor = 22 [n = 89]: weight: 0.000 loss: 1.000
location_date_sensor = 23 [n = 11]: weight: 0.000 loss: 1.000
location_date_sensor = 24 [n = 4]: weight: 0.000 loss: 1.000
location_date_sensor = 25 [n = 7]: weight: 0.000 loss: 1.000
location_date_sensor = 26 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 27 [n = 17]: weight: 0.000 loss: 1.000
location_date_sensor = 28 [n = 14]: weight: 0.000 loss: 1.000
location_date_sensor = 29 [n = 8]: weight: 0.000 loss: 1.000
location_date_sensor = 30 [n = 43]: weight: 0.000 loss: 1.000
location_date_sensor = 31 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 32 [n = 51]: weight: 0.000 loss: 1.000
location_date_sensor = 33 [n = 50]: weight: 0.000 loss: 1.000
location_date_sensor = 34 [n = 28]: weight: 0.000 loss: 1.000
location_date_sensor = 35 [n = 75]: weight: 0.000 loss: 1.000
location_date_sensor = 36 [n = 56]: weight: 0.000 loss: 1.000
location_date_sensor = 37 [n = 34]: weight: 0.000 loss: 1.000
location_date_sensor = 38 [n = 55]: weight: 0.000 loss: 1.000
location_date_sensor = 39 [n = 13]: weight: 0.000 loss: 1.000
location_date_sensor = 40 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 41 [n = 19]: weight: 0.000 loss: 1.000
location_date_sensor = 42 [n = 57]: weight: 0.000 loss: 1.000
location_date_sensor = 43 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 44 [n = 33]: weight: 0.000 loss: 1.000
location_date_sensor = 45 [n = 39]: weight: 0.000 loss: 1.000
location_date_sensor = 46 [n = 30]: weight: 0.000 loss: 1.000
Epoch eval:
Average detection_acc: 0.016
location_date_sensor = 18 [n = 12]: detection_acc = 0.000
location_date_sensor = 19 [n = 49]: detection_acc = 0.020
location_date_sensor = 20 [n = 254]: detection_acc = 0.000
location_date_sensor = 21 [n = 216]: detection_acc = 0.000
location_date_sensor = 22 [n = 89]: detection_acc = 0.000
location_date_sensor = 23 [n = 11]: detection_acc = 0.091
location_date_sensor = 24 [n = 4]: detection_acc = 0.750
location_date_sensor = 25 [n = 7]: detection_acc = 0.286
location_date_sensor = 26 [n = 14]: detection_acc = 0.071
location_date_sensor = 27 [n = 17]: detection_acc = 0.059
location_date_sensor = 28 [n = 14]: detection_acc = 0.000
location_date_sensor = 29 [n = 8]: detection_acc = 0.000
location_date_sensor = 30 [n = 43]: detection_acc = 0.000
location_date_sensor = 31 [n = 55]: detection_acc = 0.000
location_date_sensor = 32 [n = 51]: detection_acc = 0.000
location_date_sensor = 33 [n = 50]: detection_acc = 0.000
location_date_sensor = 34 [n = 28]: detection_acc = 0.000
location_date_sensor = 35 [n = 75]: detection_acc = 0.040
location_date_sensor = 36 [n = 56]: detection_acc = 0.071
location_date_sensor = 37 [n = 34]: detection_acc = 0.000
location_date_sensor = 38 [n = 55]: detection_acc = 0.000
location_date_sensor = 39 [n = 13]: detection_acc = 0.000
location_date_sensor = 40 [n = 19]: detection_acc = 0.000
location_date_sensor = 41 [n = 19]: detection_acc = 0.000
location_date_sensor = 42 [n = 57]: detection_acc = 0.000
location_date_sensor = 43 [n = 39]: detection_acc = 0.000
location_date_sensor = 44 [n = 33]: detection_acc = 0.030
location_date_sensor = 45 [n = 39]: detection_acc = 0.026
location_date_sensor = 46 [n = 30]: detection_acc = 0.133
Worst-group detection_acc: 0.000
Validation detection_acc_avg: 0.016
Epoch 3 has the best validation performance so far.
100% 365/365 [03:29<00:00, 1.74it/s]
Epoch [4]:
Train:
5% 49/915 [01:15<22:04, 1.53s/it]objective: 0.134
loss_avg: 0.634
location_date_sensor = 0 [n = 12]: weight: 0.045 loss: 0.524
location_date_sensor = 1 [n = 10]: weight: 0.007 loss: 0.519
location_date_sensor = 2 [n = 16]: weight: 0.025 loss: 0.598
location_date_sensor = 3 [n = 7]: weight: 0.023 loss: 0.516
location_date_sensor = 4 [n = 9]: weight: 0.036 loss: 0.557
location_date_sensor = 5 [n = 7]: weight: 0.016 loss: 0.615
location_date_sensor = 6 [n = 11]: weight: 0.034 loss: 0.752
location_date_sensor = 7 [n = 14]: weight: 0.088 loss: 0.575
location_date_sensor = 8 [n = 15]: weight: 0.032 loss: 0.618
location_date_sensor = 9 [n = 11]: weight: 0.083 loss: 0.685
location_date_sensor = 10 [n = 9]: weight: 0.116 loss: 0.749
location_date_sensor = 11 [n = 12]: weight: 0.042 loss: 0.575
location_date_sensor = 12 [n = 10]: weight: 0.036 loss: 0.732
location_date_sensor = 13 [n = 16]: weight: 0.112 loss: 0.633
location_date_sensor = 14 [n = 7]: weight: 0.100 loss: 0.877
location_date_sensor = 15 [n = 9]: weight: 0.147 loss: 0.844
location_date_sensor = 16 [n = 8]: weight: 0.043 loss: 0.635
location_date_sensor = 17 [n = 17]: weight: 0.018 loss: 0.573
11% 99/915 [02:31<20:45, 1.53s/it]objective: 0.135
loss_avg: 0.640
location_date_sensor = 0 [n = 16]: weight: 0.045 loss: 0.666
location_date_sensor = 1 [n = 17]: weight: 0.007 loss: 0.508
location_date_sensor = 2 [n = 13]: weight: 0.025 loss: 0.535
location_date_sensor = 3 [n = 5]: weight: 0.022 loss: 0.532
location_date_sensor = 4 [n = 10]: weight: 0.036 loss: 0.556
location_date_sensor = 5 [n = 13]: weight: 0.016 loss: 0.633
location_date_sensor = 6 [n = 3]: weight: 0.034 loss: 0.673
location_date_sensor = 7 [n = 9]: weight: 0.088 loss: 0.590
location_date_sensor = 8 [n = 16]: weight: 0.032 loss: 0.588
location_date_sensor = 9 [n = 12]: weight: 0.083 loss: 0.683
location_date_sensor = 10 [n = 15]: weight: 0.118 loss: 0.886
location_date_sensor = 11 [n = 9]: weight: 0.041 loss: 0.583
location_date_sensor = 12 [n = 9]: weight: 0.036 loss: 0.655
location_date_sensor = 13 [n = 9]: weight: 0.111 loss: 0.586
location_date_sensor = 14 [n = 12]: weight: 0.101 loss: 0.776
location_date_sensor = 15 [n = 10]: weight: 0.147 loss: 0.816
location_date_sensor = 16 [n = 13]: weight: 0.043 loss: 0.628
location_date_sensor = 17 [n = 9]: weight: 0.018 loss: 0.560
16% 149/915 [03:48<19:35, 1.53s/it]objective: 0.133
loss_avg: 0.643
location_date_sensor = 0 [n = 16]: weight: 0.045 loss: 0.554
location_date_sensor = 1 [n = 12]: weight: 0.007 loss: 0.546
location_date_sensor = 2 [n = 13]: weight: 0.025 loss: 0.566
location_date_sensor = 3 [n = 10]: weight: 0.022 loss: 0.540
location_date_sensor = 4 [n = 9]: weight: 0.035 loss: 0.619
location_date_sensor = 5 [n = 13]: weight: 0.016 loss: 0.588
location_date_sensor = 6 [n = 14]: weight: 0.032 loss: 0.687
location_date_sensor = 7 [n = 7]: weight: 0.084 loss: 0.549
location_date_sensor = 8 [n = 9]: weight: 0.033 loss: 0.599
location_date_sensor = 9 [n = 6]: weight: 0.082 loss: 0.680
location_date_sensor = 10 [n = 10]: weight: 0.121 loss: 0.809
location_date_sensor = 11 [n = 16]: weight: 0.040 loss: 0.520
location_date_sensor = 12 [n = 14]: weight: 0.036 loss: 0.716
location_date_sensor = 13 [n = 7]: weight: 0.109 loss: 0.779
location_date_sensor = 14 [n = 15]: weight: 0.106 loss: 0.853
location_date_sensor = 15 [n = 10]: weight: 0.147 loss: 0.778
location_date_sensor = 16 [n = 9]: weight: 0.042 loss: 0.697
location_date_sensor = 17 [n = 10]: weight: 0.018 loss: 0.557
22% 199/915 [05:04<18:16, 1.53s/it]objective: 0.139
loss_avg: 0.637
location_date_sensor = 0 [n = 10]: weight: 0.045 loss: 0.488
location_date_sensor = 1 [n = 14]: weight: 0.007 loss: 0.498
location_date_sensor = 2 [n = 11]: weight: 0.025 loss: 0.614
location_date_sensor = 3 [n = 8]: weight: 0.021 loss: 0.580
location_date_sensor = 4 [n = 12]: weight: 0.035 loss: 0.594
location_date_sensor = 5 [n = 11]: weight: 0.016 loss: 0.668
location_date_sensor = 6 [n = 8]: weight: 0.032 loss: 0.669
location_date_sensor = 7 [n = 11]: weight: 0.082 loss: 0.550
location_date_sensor = 8 [n = 13]: weight: 0.032 loss: 0.603
location_date_sensor = 9 [n = 6]: weight: 0.078 loss: 0.659
location_date_sensor = 10 [n = 8]: weight: 0.120 loss: 0.719
location_date_sensor = 11 [n = 11]: weight: 0.040 loss: 0.515
location_date_sensor = 12 [n = 13]: weight: 0.037 loss: 0.678
location_date_sensor = 13 [n = 17]: weight: 0.112 loss: 0.641
location_date_sensor = 14 [n = 13]: weight: 0.109 loss: 0.838
location_date_sensor = 15 [n = 14]: weight: 0.150 loss: 0.798
location_date_sensor = 16 [n = 7]: weight: 0.042 loss: 0.807
location_date_sensor = 17 [n = 13]: weight: 0.018 loss: 0.581
27% 249/915 [06:21<17:00, 1.53s/it]objective: 0.155
loss_avg: 0.647
location_date_sensor = 0 [n = 9]: weight: 0.045 loss: 0.515
location_date_sensor = 1 [n = 4]: weight: 0.007 loss: 0.534
location_date_sensor = 2 [n = 11]: weight: 0.024 loss: 0.615
location_date_sensor = 3 [n = 10]: weight: 0.021 loss: 0.579
location_date_sensor = 4 [n = 13]: weight: 0.035 loss: 0.615
location_date_sensor = 5 [n = 13]: weight: 0.016 loss: 0.541
location_date_sensor = 6 [n = 10]: weight: 0.031 loss: 0.636
location_date_sensor = 7 [n = 12]: weight: 0.082 loss: 0.614
location_date_sensor = 8 [n = 14]: weight: 0.032 loss: 0.593
location_date_sensor = 9 [n = 9]: weight: 0.078 loss: 0.703
location_date_sensor = 10 [n = 15]: weight: 0.122 loss: 0.787
location_date_sensor = 11 [n = 14]: weight: 0.040 loss: 0.561
location_date_sensor = 12 [n = 8]: weight: 0.036 loss: 0.699
location_date_sensor = 13 [n = 10]: weight: 0.111 loss: 0.604
location_date_sensor = 14 [n = 16]: weight: 0.113 loss: 0.852
location_date_sensor = 15 [n = 12]: weight: 0.150 loss: 0.760
location_date_sensor = 16 [n = 7]: weight: 0.041 loss: 0.725
location_date_sensor = 17 [n = 13]: weight: 0.018 loss: 0.577
33% 299/915 [07:37<15:38, 1.52s/it]objective: 0.141
loss_avg: 0.636
location_date_sensor = 0 [n = 15]: weight: 0.044 loss: 0.511
location_date_sensor = 1 [n = 11]: weight: 0.006 loss: 0.504
location_date_sensor = 2 [n = 12]: weight: 0.024 loss: 0.590
location_date_sensor = 3 [n = 11]: weight: 0.020 loss: 0.516
location_date_sensor = 4 [n = 9]: weight: 0.034 loss: 0.581
location_date_sensor = 5 [n = 14]: weight: 0.016 loss: 0.751
location_date_sensor = 6 [n = 11]: weight: 0.031 loss: 0.625
location_date_sensor = 7 [n = 14]: weight: 0.081 loss: 0.576
location_date_sensor = 8 [n = 4]: weight: 0.031 loss: 0.634
location_date_sensor = 9 [n = 12]: weight: 0.077 loss: 0.618
location_date_sensor = 10 [n = 13]: weight: 0.124 loss: 0.749
location_date_sensor = 11 [n = 10]: weight: 0.040 loss: 0.553
location_date_sensor = 12 [n = 15]: weight: 0.036 loss: 0.701
location_date_sensor = 13 [n = 11]: weight: 0.110 loss: 0.624
location_date_sensor = 14 [n = 11]: weight: 0.117 loss: 0.861
location_date_sensor = 15 [n = 10]: weight: 0.152 loss: 0.759
location_date_sensor = 16 [n = 7]: weight: 0.040 loss: 0.721
location_date_sensor = 17 [n = 10]: weight: 0.018 loss: 0.587
38% 349/915 [08:54<14:28, 1.53s/it]objective: 0.126
loss_avg: 0.632
location_date_sensor = 0 [n = 13]: weight: 0.044 loss: 0.529
location_date_sensor = 1 [n = 11]: weight: 0.006 loss: 0.514
location_date_sensor = 2 [n = 14]: weight: 0.024 loss: 0.593
location_date_sensor = 3 [n = 15]: weight: 0.020 loss: 0.553
location_date_sensor = 4 [n = 11]: weight: 0.034 loss: 0.574
location_date_sensor = 5 [n = 11]: weight: 0.016 loss: 0.658
location_date_sensor = 6 [n = 13]: weight: 0.032 loss: 0.679
location_date_sensor = 7 [n = 9]: weight: 0.082 loss: 0.625
location_date_sensor = 8 [n = 9]: weight: 0.031 loss: 0.628
location_date_sensor = 9 [n = 12]: weight: 0.078 loss: 0.626
location_date_sensor = 10 [n = 8]: weight: 0.125 loss: 0.764
location_date_sensor = 11 [n = 11]: weight: 0.039 loss: 0.512
location_date_sensor = 12 [n = 16]: weight: 0.038 loss: 0.705
location_date_sensor = 13 [n = 8]: weight: 0.108 loss: 0.606
location_date_sensor = 14 [n = 9]: weight: 0.119 loss: 0.918
location_date_sensor = 15 [n = 6]: weight: 0.150 loss: 0.816
location_date_sensor = 16 [n = 11]: weight: 0.040 loss: 0.647
location_date_sensor = 17 [n = 13]: weight: 0.017 loss: 0.607
44% 399/915 [10:10<13:09, 1.53s/it]objective: 0.147
loss_avg: 0.660
location_date_sensor = 0 [n = 13]: weight: 0.044 loss: 0.515
location_date_sensor = 1 [n = 10]: weight: 0.006 loss: 0.551
location_date_sensor = 2 [n = 12]: weight: 0.024 loss: 0.558
location_date_sensor = 3 [n = 9]: weight: 0.020 loss: 0.560
location_date_sensor = 4 [n = 10]: weight: 0.033 loss: 0.573
location_date_sensor = 5 [n = 15]: weight: 0.017 loss: 0.794
location_date_sensor = 6 [n = 8]: weight: 0.032 loss: 0.640
location_date_sensor = 7 [n = 10]: weight: 0.081 loss: 0.538
location_date_sensor = 8 [n = 10]: weight: 0.030 loss: 0.633
location_date_sensor = 9 [n = 7]: weight: 0.077 loss: 0.704
location_date_sensor = 10 [n = 9]: weight: 0.124 loss: 0.813
location_date_sensor = 11 [n = 9]: weight: 0.038 loss: 0.533
location_date_sensor = 12 [n = 17]: weight: 0.040 loss: 0.746
location_date_sensor = 13 [n = 14]: weight: 0.109 loss: 0.686
location_date_sensor = 14 [n = 15]: weight: 0.121 loss: 0.805
location_date_sensor = 15 [n = 14]: weight: 0.150 loss: 0.800
location_date_sensor = 16 [n = 8]: weight: 0.039 loss: 0.546
location_date_sensor = 17 [n = 10]: weight: 0.018 loss: 0.657
49% 449/915 [11:26<11:51, 1.53s/it]objective: 0.139
loss_avg: 0.652
location_date_sensor = 0 [n = 9]: weight: 0.044 loss: 0.550
location_date_sensor = 1 [n = 6]: weight: 0.006 loss: 0.551
location_date_sensor = 2 [n = 12]: weight: 0.024 loss: 0.660
location_date_sensor = 3 [n = 10]: weight: 0.020 loss: 0.553
location_date_sensor = 4 [n = 10]: weight: 0.032 loss: 0.577
location_date_sensor = 5 [n = 12]: weight: 0.017 loss: 0.701
location_date_sensor = 6 [n = 17]: weight: 0.032 loss: 0.762
location_date_sensor = 7 [n = 15]: weight: 0.080 loss: 0.548
location_date_sensor = 8 [n = 11]: weight: 0.030 loss: 0.598
location_date_sensor = 9 [n = 9]: weight: 0.077 loss: 0.645
location_date_sensor = 10 [n = 13]: weight: 0.125 loss: 0.753
location_date_sensor = 11 [n = 12]: weight: 0.038 loss: 0.505
location_date_sensor = 12 [n = 10]: weight: 0.040 loss: 0.684
location_date_sensor = 13 [n = 12]: weight: 0.111 loss: 0.643
location_date_sensor = 14 [n = 8]: weight: 0.120 loss: 0.896
location_date_sensor = 15 [n = 7]: weight: 0.152 loss: 0.853
location_date_sensor = 16 [n = 12]: weight: 0.039 loss: 0.623
location_date_sensor = 17 [n = 15]: weight: 0.017 loss: 0.659
55% 499/915 [12:43<10:35, 1.53s/it]objective: 0.147
loss_avg: 0.636
location_date_sensor = 0 [n = 8]: weight: 0.042 loss: 0.556
location_date_sensor = 1 [n = 13]: weight: 0.006 loss: 0.510
location_date_sensor = 2 [n = 12]: weight: 0.024 loss: 0.546
location_date_sensor = 3 [n = 9]: weight: 0.019 loss: 0.570
location_date_sensor = 4 [n = 11]: weight: 0.032 loss: 0.590
location_date_sensor = 5 [n = 11]: weight: 0.017 loss: 0.688
location_date_sensor = 6 [n = 7]: weight: 0.033 loss: 0.739
location_date_sensor = 7 [n = 6]: weight: 0.079 loss: 0.659
location_date_sensor = 8 [n = 14]: weight: 0.030 loss: 0.617
location_date_sensor = 9 [n = 12]: weight: 0.076 loss: 0.668
location_date_sensor = 10 [n = 8]: weight: 0.126 loss: 0.777
location_date_sensor = 11 [n = 14]: weight: 0.037 loss: 0.545
location_date_sensor = 12 [n = 10]: weight: 0.040 loss: 0.688
location_date_sensor = 13 [n = 14]: weight: 0.110 loss: 0.611
location_date_sensor = 14 [n = 6]: weight: 0.118 loss: 0.805
location_date_sensor = 15 [n = 18]: weight: 0.156 loss: 0.816
location_date_sensor = 16 [n = 14]: weight: 0.039 loss: 0.610
location_date_sensor = 17 [n = 13]: weight: 0.017 loss: 0.564
60% 549/915 [13:59<09:20, 1.53s/it]objective: 0.141
loss_avg: 0.647
location_date_sensor = 0 [n = 11]: weight: 0.041 loss: 0.535
location_date_sensor = 1 [n = 9]: weight: 0.006 loss: 0.514
location_date_sensor = 2 [n = 11]: weight: 0.023 loss: 0.614
location_date_sensor = 3 [n = 10]: weight: 0.019 loss: 0.545
location_date_sensor = 4 [n = 6]: weight: 0.031 loss: 0.587
location_date_sensor = 5 [n = 12]: weight: 0.017 loss: 0.705
location_date_sensor = 6 [n = 14]: weight: 0.032 loss: 0.700
location_date_sensor = 7 [n = 12]: weight: 0.078 loss: 0.633
location_date_sensor = 8 [n = 15]: weight: 0.030 loss: 0.586
location_date_sensor = 9 [n = 8]: weight: 0.074 loss: 0.645
location_date_sensor = 10 [n = 7]: weight: 0.125 loss: 0.689
location_date_sensor = 11 [n = 9]: weight: 0.037 loss: 0.540
location_date_sensor = 12 [n = 14]: weight: 0.040 loss: 0.672
location_date_sensor = 13 [n = 12]: weight: 0.111 loss: 0.633
location_date_sensor = 14 [n = 16]: weight: 0.121 loss: 0.815
location_date_sensor = 15 [n = 8]: weight: 0.162 loss: 0.861
location_date_sensor = 16 [n = 12]: weight: 0.038 loss: 0.690
location_date_sensor = 17 [n = 14]: weight: 0.017 loss: 0.602
65% 599/915 [15:16<08:02, 1.53s/it]objective: 0.141
loss_avg: 0.629
location_date_sensor = 0 [n = 9]: weight: 0.041 loss: 0.464
location_date_sensor = 1 [n = 14]: weight: 0.006 loss: 0.504
location_date_sensor = 2 [n = 7]: weight: 0.023 loss: 0.562
location_date_sensor = 3 [n = 7]: weight: 0.018 loss: 0.552
location_date_sensor = 4 [n = 13]: weight: 0.031 loss: 0.559
location_date_sensor = 5 [n = 14]: weight: 0.017 loss: 0.613
location_date_sensor = 6 [n = 10]: weight: 0.032 loss: 0.670
location_date_sensor = 7 [n = 11]: weight: 0.077 loss: 0.541
location_date_sensor = 8 [n = 10]: weight: 0.030 loss: 0.574
location_date_sensor = 9 [n = 4]: weight: 0.071 loss: 0.570
location_date_sensor = 10 [n = 14]: weight: 0.123 loss: 0.682
location_date_sensor = 11 [n = 12]: weight: 0.036 loss: 0.548
location_date_sensor = 12 [n = 12]: weight: 0.039 loss: 0.697
location_date_sensor = 13 [n = 11]: weight: 0.110 loss: 0.626
location_date_sensor = 14 [n = 10]: weight: 0.126 loss: 0.868
location_date_sensor = 15 [n = 13]: weight: 0.164 loss: 0.892
location_date_sensor = 16 [n = 13]: weight: 0.039 loss: 0.651
location_date_sensor = 17 [n = 16]: weight: 0.018 loss: 0.637
67% 612/915 [15:36<07:43, 1.53s/it]
Preparing a submission¶
The WILDS library saves predictions in *_pred.pth files. To prepare them for submission to our competition, we have to transform them into the appropriate format.
In [ ]:
import torch
import pandas as pd
pred_false_val = torch.load(f"{SAVE_TO}/gwhd_split:val_seed:{N_EPOCHS-1}_epoch:{N_EPOCHS-1}_pred.pth")
pred_false_test = torch.load(f"{SAVE_TO}/gwhd_split:test_seed:{N_EPOCHS-1}_epoch:{N_EPOCHS-1}_pred.pth")
false_val = pd.read_csv("data/gwhd_v0.9/official_val.csv")
false_test = pd.read_csv("data/gwhd_v0.9/official_test.csv")
In [ ]:
assert len(false_val) == len(pred_false_val)
assert len(pred_false_test) == len(pred_false_test)
In [ ]:
def encodeBoxes(tensor_boxes,score_thr=0.5):
boxes = tensor_boxes["boxes"].numpy()
scores = tensor_boxes["scores"].numpy()
boxes = boxes[scores > score_thr]
strboxes = ";".join([" ".join([str(int(i)) for i in box]) for box in boxes])
if len(strboxes) == 0:
strboxes = "no_box"
return strboxes
In [ ]:
encoded_boxes = [encodeBoxes(tensor_boxes) for tensor_boxes in pred_false_val]
false_val["PredString"] = encoded_boxes
encoded_boxes = [encodeBoxes(tensor_boxes) for tensor_boxes in pred_false_test]
false_test["PredString"] = encoded_boxes
Sanity check before submission¶
We can check few predictions to get an idea of how well the training went
In [ ]:
test_img = "data/gwhd_v0.9/images/"+ false_test["image_name"].values[1]
print(test_img)
import cv2
import matplotlib.pyplot as plt
test_img = cv2.imread(test_img)
boxes = pred_false_test[1]["boxes"].numpy()
scores = pred_false_test[1]["scores"].numpy()
boxes = boxes[scores >0.5] # we set a naive threshold here
for (x,y,xx,yy) in boxes:
cv2.rectangle(test_img,(int(x),int(y)),(int(xx),int(yy)),(255,0,0),5)
plt.imshow(test_img[...,::-1])
Writing submission file¶
In [ ]:
results = pd.concat([false_test,false_val]) #Val and test does not correspond to the true split
del results["BoxesString"] # We need to remove this column or it causes damage :)
results["image_name"] = results["image_name"].apply(lambda x: x.replace(".png","")) # we need to remove the extension
results.to_csv("submission_final.csv",index=False)
Making Direct Submission thought Aicrowd CLI¶
In [ ]:
!aicrowd submission create -c global-wheat-challenge-2021 -f submission_final.csv
In [ ]:
Content
Comments
You must login before you can post a comment.