Files
pyannote-audio/tutorials/voice_activity_detection.ipynb

2 lines
138 KiB
Plaintext
Raw Permalink Blame History

This file contains invisible Unicode characters
This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
{"cells":[{"cell_type":"markdown","metadata":{"id":"3M4zoYI4Q-Xi"},"source":["<a href=\"https://colab.research.google.com/github/pyannote/pyannote-audio/blob/develop/tutorials/voice_activity_detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"]},{"cell_type":"markdown","metadata":{"id":"fEew3B3-Q-Xl"},"source":["# Training a voice activity detection pipeline from scratch with `pyannote.audio`\n","\n","Voice activity detection (VAD) is the task of detecting speech regions in a given audio stream or recording. \n","In this tutorial, you will learn how to train and evaluate a VAD pipeline from scratch."]},{"cell_type":"markdown","metadata":{"id":"1F6PKPftQ-Xm"},"source":["## Tutorial setup"]},{"cell_type":"markdown","metadata":{"id":"ErPKDILdQ-Xn"},"source":["### `Google Colab` setup"]},{"cell_type":"markdown","metadata":{"id":"SYU2JJ9-Q-Xo"},"source":["If you are running this tutorial on `Colab`, execute the following commands in order to setup `Colab` environment. These commands will install `pyannote.audio` and download a mini version of the `AMI` corpus."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":1772401,"status":"ok","timestamp":1704890790335,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"EMspGag4Q-Xp","outputId":"0aaf740d-96c8-4340-a9e7-5c529f12fe9b"},"outputs":[],"source":["!pip install -qq pyannote.audio==3.1.1\n","!pip install -qq ipython==7.34.0\n","!git clone https://github.com/pyannote/AMI-diarization-setup.git\n","%cd ./AMI-diarization-setup/pyannote/\n","!bash ./download_ami_mini.sh\n","%cd /content"]},{"cell_type":"markdown","metadata":{"id":"TFqu_5vyQ-Xq"},"source":["⚠ Restart the runtime (Runtime > Restart session)."]},{"cell_type":"markdown","metadata":{"id":"7_7eodg5Q-Xr"},"source":["### Non `Google Colab` setup"]},{"cell_type":"markdown","metadata":{"id":"WCy8p55QQ-Xr"},"source":["If you are not using `Colab`, this tutorial assumes that\n","* `pyannote.audio` has been installed\n","* the [AMI corpus](https://groups.inf.ed.ac.uk/ami/corpus/) has already been [setup for use with `pyannote`](https://github.com/pyannote/AMI-diarization-setup/tree/main/pyannote)"]},{"cell_type":"markdown","metadata":{"id":"M-oe_8FFQ-Xu"},"source":["## Data preparation\n","\n","See [`pyannote.database` documentation](https://github.com/pyannote/pyannote-database#pyannote-database) to learn how to prepare your own dataset for training with `pyannote.audio`."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"KBstX_2sQ-Xu"},"outputs":[],"source":["from pyannote.database import registry, FileFinder\n","\n","registry.load_database(\"AMI-diarization-setup/pyannote/database.yml\")\n","preprocessors = {\"audio\": FileFinder()}\n","ami = registry.get_protocol('AMI.SpeakerDiarization.mini',\n"," preprocessors=preprocessors)"]},{"cell_type":"markdown","metadata":{"id":"oYIKQ5kwQ-Xv"},"source":["`pyannote.database` *protocols* usually define\n","* a training set: `for training_file in protocol.train(): ...`,\n","* a validation set: `for validation_file in protocol.development(): ...`\n","* an evaluation set `for evaluation_file in protocol.test(): ...`"]},{"cell_type":"markdown","metadata":{"id":"siLd6sE3Q-Xw"},"source":["Let's listen to one minute of the first training file..."]},{"cell_type":"code","execution_count":2,"metadata":{"executionInfo":{"elapsed":212,"status":"ok","timestamp":1704890828290,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"bai4OplyQ-Xw"},"outputs":[],"source":["first_training_file = next(ami.train())\n","reference = first_training_file[\"annotation\"]"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":151,"output_embedded_package_id":"1GxUBhwcdeeXdLogFqdzXqSDl6FNpaR2t"},"executionInfo":{"elapsed":7227,"status":"ok","timestamp":1704890839349,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"5t2l6o3NQ-Xx","outputId":"9755f212-b8ca-45ad-e1a2-60f00665d7be"},"outputs":[],"source":["from pyannote.audio.utils.preview import listen\n","from pyannote.core import Segment\n","\n","one_minute = Segment(240, 300)\n","listen(first_training_file, one_minute)"]},{"cell_type":"markdown","metadata":{"id":"Z3-VQas0Q-Xy"},"source":["... and visualize the manual annotation:"]},{"cell_type":"code","execution_count":4,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":239},"executionInfo":{"elapsed":592,"status":"ok","timestamp":1704890847735,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"8fNi_I0pQ-Xz","outputId":"83cfee16-8a4e-4528-8a60-d5b8c0990fc7"},"outputs":[{"data":{"image/png":"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","text/plain":["<pyannote.core.annotation.Annotation at 0x7fa750ac2410>"]},"execution_count":4,"metadata":{},"output_type":"execute_result"}],"source":["from pyannote.core import notebook\n","notebook.crop = one_minute\n","reference"]},{"cell_type":"markdown","metadata":{"id":"fzbxnisZQ-Xz"},"source":["## Training"]},{"cell_type":"markdown","metadata":{"id":"C_QEjNxjQ-Xz"},"source":["We initialize a VAD *task* that describes how the model will be trained:\n","\n","* `ami` indicates that we will use files available in `ami.train()`.\n","* `duration=2.` and `batch_size=128` indicates that the model will ingest batches of 128 two seconds long audio chunks."]},{"cell_type":"code","execution_count":5,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":543,"status":"ok","timestamp":1704890854558,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"oRkU-MboQ-Xz","outputId":"173a3d4b-fdf4-4e13-817d-dc024ac87524"},"outputs":[{"name":"stdout","output_type":"stream","text":["Protocol AMI.SpeakerDiarization.mini does not precompute the output of torchaudio.info(): adding a 'torchaudio.info' preprocessor for you to speed up dataloaders. See pyannote.database documentation on how to do that yourself.\n"]}],"source":["from pyannote.audio.tasks import VoiceActivityDetection\n","\n","vad = VoiceActivityDetection(ami, duration=2., batch_size=128)"]},{"cell_type":"markdown","metadata":{"id":"A7q9pnI4Q-X1"},"source":["We initialize one *model* with the `PyanNet` architecture used [in that paper](https://arxiv.org/abs/2104.04045). \n","In particular, we increase the default stride of the initial `sincnet` feature extraction layer to `10`.\n","\n","The model is also provided with the task (`task=vad`) for which it is being trained:"]},{"cell_type":"code","execution_count":6,"metadata":{"executionInfo":{"elapsed":241,"status":"ok","timestamp":1704890861715,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"dHJMJHB6Q-X1"},"outputs":[],"source":["from pyannote.audio.models.segmentation import PyanNet\n","\n","model = PyanNet(sincnet={'stride': 10}, task=vad)"]},{"cell_type":"markdown","metadata":{"id":"Rwh2ulN2Q-X2"},"source":["Now that everything is ready, let's train with `pytorch-lightning`!"]},{"cell_type":"code","execution_count":7,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":414,"referenced_widgets":["de7bbc2d644e4ba79f3b4dcc26da3cb7","b62ef7013afe4da4a00ba0a70a63e938","b20cbf0c630a44f8b108d141e0b14ecc","4946ea8187904c7fa9f7c5de92a563e2","929954f82bc6407aaa6b069da848d1e3","53ac5430d2ef45ecaaecc3da0706825b","ffd1d20386d14e2f945f2d84df7a681f","69c34a612abb4d7a83bd796ef97d1c92","2ea0e51154e54129a6b164fc72588edc","910983c75b5247deb203ab218e05b25a","a1861066848a4efda4cc1f09beacb451","321302a76845477cbfd6b644071048b9","0300ea5285a94c7580c789849ea8b7e7","51dc1507c16e4fd88da6e9681b6f8443","84560e1a1a154cf7a4cc998aaa7db862","dce43bbd970643d99f8abbcb299925ec","f3bb471baf44445d9ea50d0adde041c1","14556644c3104b02850a07f172f1411a","50f8b2a4995242ac9d9961f34f2223ae","622cf364efe34719877311f1a46d24b4","ef09269769d845fb84ba39bd43a4ee0c","b827c08a92434a31b82d0382057ee82b","d5dbc90f387a4c9eb43789707505e617","3b55a90beb874204a44df7f231da55eb","8429d57eec2f4c9c8cda3900c0da5df5","d4f4fff579354b10b3cdb32cb11754fe","b886a9d46c7647f0aa5e04260a5b6e1e","5151fae07f114d4e86ee7ec4e487faae","4d2ea6328f684e6e888bc5710dea4340","290e8084737b4ff7a816f4b67fa8d411","bb89bf63ea124306acf411cbeaca2b90","e17799ed71e04167bd46363cf84cc93f","146c185d279d4583860ba4f9defda025"]},"executionInfo":{"elapsed":1454764,"status":"ok","timestamp":1704892319974,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"KkFsUEV_Q-X2","outputId":"3361c2b5-f341-41e1-e027-8fb9b7404d34"},"outputs":[{"name":"stderr","output_type":"stream","text":["INFO:pytorch_lightning.utilities.rank_zero:GPU available: True (cuda), used: True\n","INFO:pytorch_lightning.utilities.rank_zero:TPU available: False, using: 0 TPU cores\n","INFO:pytorch_lightning.utilities.rank_zero:IPU available: False, using: 0 IPUs\n","INFO:pytorch_lightning.utilities.rank_zero:HPU available: False, using: 0 HPUs\n","WARNING:pytorch_lightning.loggers.tensorboard:Missing logger folder: /content/lightning_logs\n","INFO:pytorch_lightning.accelerators.cuda:LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n","INFO:pytorch_lightning.callbacks.model_summary:\n"," | Name | Type | Params | In sizes | Out sizes \n","---------------------------------------------------------------------------------------------------------------------\n","0 | sincnet | SincNet | 42.6 K | [1, 1, 32000] | [1, 60, 115] \n","1 | lstm | LSTM | 589 K | [1, 115, 60] | [[1, 115, 256], [[4, 1, 128], [4, 1, 128]]]\n","2 | linear | ModuleList | 49.4 K | ? | ? \n","3 | classifier | Linear | 129 | [1, 115, 128] | [1, 115, 1] \n","4 | activation | Sigmoid | 0 | [1, 115, 1] | [1, 115, 1] \n","5 | validation_metric | MetricCollection | 0 | ? | ? \n","---------------------------------------------------------------------------------------------------------------------\n","681 K Trainable params\n","0 Non-trainable params\n","681 K Total params\n","2.728 Total estimated model params size (MB)\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"de7bbc2d644e4ba79f3b4dcc26da3cb7","version_major":2,"version_minor":0},"text/plain":["Sanity Checking: | | 0/? [00:00<?, ?it/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"321302a76845477cbfd6b644071048b9","version_major":2,"version_minor":0},"text/plain":["Training: | | 0/? [00:00<?, ?it/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"d5dbc90f387a4c9eb43789707505e617","version_major":2,"version_minor":0},"text/plain":["Validation: | | 0/? [00:00<?, ?it/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stderr","output_type":"stream","text":["INFO:pytorch_lightning.utilities.rank_zero:`Trainer.fit` stopped: `max_epochs=1` reached.\n"]}],"source":["import pytorch_lightning as pl\n","\n","trainer = pl.Trainer(devices=1, max_epochs=1)\n","\n","trainer.fit(model)"]},{"cell_type":"markdown","metadata":{"id":"1yEg2_iaQ-X3"},"source":["For the purpose of this tutorial, the model is trained for only 1 epoch. One can obviously expect better performance by training longer and on more data.\n","\n","See [`pytorch-lightning`](https://www.pytorchlightning.ai/) documentation to learn more about the [`Trainer` API](https://pytorch-lightning.readthedocs.io/en/latest/common/trainer.html), in particular."]},{"cell_type":"markdown","metadata":{"id":"nc_7Gwq3Q-X3"},"source":["## Inference\n","\n","Once trained, the model can be applied to a test file:"]},{"cell_type":"code","execution_count":8,"metadata":{"executionInfo":{"elapsed":243,"status":"ok","timestamp":1704892548416,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"KNLZmquRQ-X4"},"outputs":[],"source":["test_file = next(ami.test())\n","# here we use a test file provided by the protocol, but it could be any audio file\n","# e.g. test_file = \"/path/to/test.wav\"."]},{"cell_type":"markdown","metadata":{"id":"rli7ni6qQ-X4"},"source":["Because the model was trained on 2s audio chunks and that test files are likely to be much longer than that, we wrap the `model` with an `Inference` instance: it will take care of sliding a 2s window over the whole file and aggregate the output of the model."]},{"cell_type":"code","execution_count":9,"metadata":{"executionInfo":{"elapsed":41494,"status":"ok","timestamp":1704892592044,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"3te_7h_vQ-X5"},"outputs":[],"source":["from pyannote.audio import Inference\n","inference = Inference(model)\n","vad_probability = inference(test_file)"]},{"cell_type":"code","execution_count":10,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":216},"executionInfo":{"elapsed":385,"status":"ok","timestamp":1704892629148,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"-pPasQhzQ-X5","outputId":"43867a67-db1c-4292-c326-c923aa5ef2c9"},"outputs":[{"data":{"image/png":"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","text/plain":["<pyannote.core.feature.SlidingWindowFeature at 0x7fa721f3f430>"]},"execution_count":10,"metadata":{},"output_type":"execute_result"}],"source":["vad_probability"]},{"cell_type":"markdown","metadata":{"id":"Mscb9fDsQ-X6"},"source":["Perfect voice activity detection output should look like that:"]},{"cell_type":"code","execution_count":11,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":145},"executionInfo":{"elapsed":689,"status":"ok","timestamp":1704892633130,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"ObtOlbOTQ-X7","outputId":"8dd51edd-e56e-4ae1-e95c-5a592c1a106c"},"outputs":[{"data":{"image/png":"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","text/plain":["<Timeline(uri=IS1009a, segments=[<Segment(54.95, 60.85)>, <Segment(61.54, 80.83)>, <Segment(81.04, 82.05)>, <Segment(83.39, 86.94)>, <Segment(87.49, 88.59)>, <Segment(88.82, 89.69)>, <Segment(90.21, 90.52)>, <Segment(91.31, 107.4)>, <Segment(107.88, 118.17)>, <Segment(118.76, 119.05)>, <Segment(119.38, 119.91)>, <Segment(122.4, 132.39)>, <Segment(132.87, 146.77)>, <Segment(149.44, 162.32)>, <Segment(165.19, 166.99)>, <Segment(188.06, 188.34)>, <Segment(195.04, 195.27)>, <Segment(195.56, 195.82)>, <Segment(196.38, 202.4)>, <Segment(203.78, 218.2)>, <Segment(219.44, 222.12)>, <Segment(222.76, 227.27)>, <Segment(236.11, 238.38)>, <Segment(241.01, 242.5)>, <Segment(243.94, 245.45)>, <Segment(248.32, 251.04)>, <Segment(251.8, 252.7)>, <Segment(252.79, 254.69)>, <Segment(255.24, 257.74)>, <Segment(260.02, 272.52)>, <Segment(274.68, 275.95)>, <Segment(276.23, 279.86)>, <Segment(280.08, 285.51)>, <Segment(288.1, 297.75)>, <Segment(308.92, 310.65)>, <Segment(311.19, 318.45)>, <Segment(320.53, 327.5)>, <Segment(328.77, 333.36)>, <Segment(333.57, 335.49)>, <Segment(336.02, 336.43)>, <Segment(337.47, 339.63)>, <Segment(347.67, 349.78)>, <Segment(355.86, 364.82)>, <Segment(367.25, 375.93)>, <Segment(377.25, 386.09)>, <Segment(386.53, 387.02)>, <Segment(387.4, 393.64)>, <Segment(395.23, 398.51)>, <Segment(398.76, 399.29)>, <Segment(400.24, 400.93)>, <Segment(402.86, 412.67)>, <Segment(419.16, 427.58)>, <Segment(428.69, 434.99)>, <Segment(436.56, 449.23)>, <Segment(450.37, 495.68)>, <Segment(497.24, 502.27)>, <Segment(502.77, 503.46)>, <Segment(503.78, 506.79)>, <Segment(508.17, 551.13)>, <Segment(552.51, 565.64)>, <Segment(566.02, 581.45)>, <Segment(581.48, 637.63)>, <Segment(638.22, 641.25)>, <Segment(641.3, 661.77)>, <Segment(663.98, 665.07)>, <Segment(665.22, 683.95)>, <Segment(685.84, 703.98)>, <Segment(704.61, 711.13)>, <Segment(711.71, 715.68)>, <Segment(715.72, 744.62)>, <Segment(745.25, 750.08)>, <Segment(751.38, 758)>, <Segment(764.72, 775.95)>, <Segment(777.27, 787.09)>, <Segment(787.58, 793.95)>, <Segment(794.48, 794.68)>, <Segment(795.57, 795.88)>, <Segment(796.5, 797.81)>, <Segment(797.96, 803.68)>, <Segment(803.8, 805.72)>])>"]},"execution_count":11,"metadata":{},"output_type":"execute_result"}],"source":["expected_output = test_file[\"annotation\"].get_timeline().support()\n","expected_output"]},{"cell_type":"markdown","metadata":{"id":"j0f1oMTUQ-X7"},"source":["## Pipeline"]},{"cell_type":"markdown","metadata":{"id":"QcRNQQ_VQ-X7"},"source":["Almost there! To obtain the final speech regions, we need to apply a detection threshold. \n","For that, we rely on the voice activity detection pipeline whose hyper-parameters are set manually:\n","- `onset=0.6`: mark region as `active` when probability goes above 0.6\n","- `offset=0.4`: switch back to `inactive` when probability goes below 0.4\n","- `min_duration_on=0.0`: remove `active` regions shorter than that many seconds\n","- `min_duration_off=0.0`: fill `inactive` regions shorter than that many seconds\n","\n","\n","More details about those hyper-parameters are provided in Figure 2 of [that paper](https://arxiv.org/abs/2104.04045)."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":589,"status":"ok","timestamp":1704892638091,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"TGkXstVpQ-X8","outputId":"d77d4cbe-98e9-4439-9c82-a51d0aeb4e25"},"outputs":[],"source":["from pyannote.audio.pipelines import VoiceActivityDetection as VoiceActivityDetectionPipeline\n","pipeline = VoiceActivityDetectionPipeline(segmentation=model)\n","initial_params = {\"onset\": 0.6, \"offset\": 0.4,\n"," \"min_duration_on\": 0.0, \"min_duration_off\": 0.0}\n","pipeline.instantiate(initial_params)"]},{"cell_type":"markdown","metadata":{"id":"thmmfIyjQ-X8"},"source":["Here we go:"]},{"cell_type":"code","execution_count":13,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":145},"executionInfo":{"elapsed":44782,"status":"ok","timestamp":1704892685345,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"lq9gPG2hQ-X9","outputId":"928eea8e-737c-4689-cdb7-262a71dd6b3f"},"outputs":[{"data":{"image/png":"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","text/plain":["<Timeline(uri=IS1009a, segments=[<Segment(25.7478, 30.1478)>, <Segment(32.0957, 32.4783)>, <Segment(55, 120.478)>, <Segment(122.096, 147.296)>, <Segment(149.243, 163.243)>, <Segment(164.983, 167.47)>, <Segment(195.226, 227.435)>, <Segment(236.043, 238.809)>, <Segment(243.904, 245.33)>, <Segment(248.096, 250.235)>, <Segment(251.852, 256.252)>, <Segment(257.035, 258.478)>, <Segment(259.904, 273.035)>, <Segment(274.496, 285.8)>, <Segment(288.026, 298.009)>, <Segment(307.765, 310.461)>, <Segment(310.913, 312.93)>, <Segment(315.557, 318.896)>, <Segment(320.27, 339.035)>, <Segment(347.626, 350.235)>, <Segment(357.817, 363.47)>, <Segment(367.122, 394.513)>, <Segment(395.122, 400.983)>, <Segment(402.583, 413.174)>, <Segment(418.913, 435.504)>, <Segment(436.496, 500.583)>, <Segment(502.687, 662.322)>, <Segment(663.904, 684.826)>, <Segment(685.73, 758.009)>, <Segment(764.409, 794.774)>, <Segment(795.696, 806.27)>, <Segment(815.157, 815.626)>, <Segment(821.661, 822.478)>, <Segment(825.887, 826.652)>, <Segment(828.878, 831.383)>])>"]},"execution_count":13,"metadata":{},"output_type":"execute_result"}],"source":["pipeline(test_file).get_timeline()"]},{"cell_type":"code","execution_count":14,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":145},"executionInfo":{"elapsed":237,"status":"ok","timestamp":1704892689100,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"EzQZg8VPQ-X9","outputId":"514406de-c097-4904-9b6f-a89cce4cfe89"},"outputs":[{"data":{"image/png":"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","text/plain":["<Timeline(uri=IS1009a, segments=[<Segment(54.95, 60.85)>, <Segment(61.54, 80.83)>, <Segment(81.04, 82.05)>, <Segment(83.39, 86.94)>, <Segment(87.49, 88.59)>, <Segment(88.82, 89.69)>, <Segment(90.21, 90.52)>, <Segment(91.31, 107.4)>, <Segment(107.88, 118.17)>, <Segment(118.76, 119.05)>, <Segment(119.38, 119.91)>, <Segment(122.4, 132.39)>, <Segment(132.87, 146.77)>, <Segment(149.44, 162.32)>, <Segment(165.19, 166.99)>, <Segment(188.06, 188.34)>, <Segment(195.04, 195.27)>, <Segment(195.56, 195.82)>, <Segment(196.38, 202.4)>, <Segment(203.78, 218.2)>, <Segment(219.44, 222.12)>, <Segment(222.76, 227.27)>, <Segment(236.11, 238.38)>, <Segment(241.01, 242.5)>, <Segment(243.94, 245.45)>, <Segment(248.32, 251.04)>, <Segment(251.8, 252.7)>, <Segment(252.79, 254.69)>, <Segment(255.24, 257.74)>, <Segment(260.02, 272.52)>, <Segment(274.68, 275.95)>, <Segment(276.23, 279.86)>, <Segment(280.08, 285.51)>, <Segment(288.1, 297.75)>, <Segment(308.92, 310.65)>, <Segment(311.19, 318.45)>, <Segment(320.53, 327.5)>, <Segment(328.77, 333.36)>, <Segment(333.57, 335.49)>, <Segment(336.02, 336.43)>, <Segment(337.47, 339.63)>, <Segment(347.67, 349.78)>, <Segment(355.86, 364.82)>, <Segment(367.25, 375.93)>, <Segment(377.25, 386.09)>, <Segment(386.53, 387.02)>, <Segment(387.4, 393.64)>, <Segment(395.23, 398.51)>, <Segment(398.76, 399.29)>, <Segment(400.24, 400.93)>, <Segment(402.86, 412.67)>, <Segment(419.16, 427.58)>, <Segment(428.69, 434.99)>, <Segment(436.56, 449.23)>, <Segment(450.37, 495.68)>, <Segment(497.24, 502.27)>, <Segment(502.77, 503.46)>, <Segment(503.78, 506.79)>, <Segment(508.17, 551.13)>, <Segment(552.51, 565.64)>, <Segment(566.02, 581.45)>, <Segment(581.48, 637.63)>, <Segment(638.22, 641.25)>, <Segment(641.3, 661.77)>, <Segment(663.98, 665.07)>, <Segment(665.22, 683.95)>, <Segment(685.84, 703.98)>, <Segment(704.61, 711.13)>, <Segment(711.71, 715.68)>, <Segment(715.72, 744.62)>, <Segment(745.25, 750.08)>, <Segment(751.38, 758)>, <Segment(764.72, 775.95)>, <Segment(777.27, 787.09)>, <Segment(787.58, 793.95)>, <Segment(794.48, 794.68)>, <Segment(795.57, 795.88)>, <Segment(796.5, 797.81)>, <Segment(797.96, 803.68)>, <Segment(803.8, 805.72)>])>"]},"execution_count":14,"metadata":{},"output_type":"execute_result"}],"source":["expected_output"]},{"cell_type":"markdown","metadata":{"id":"L1FOF3CcQ-X-"},"source":["Looks decent! Let's perform a proper evaluation by computing the [detection error rate](https://pyannote.github.io/pyannote-metrics/reference.html#detection) over the whole AMI test set:"]},{"cell_type":"code","execution_count":15,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":153804,"status":"ok","timestamp":1704892850039,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"6v44K7WxQ-X_","outputId":"ce83364e-147c-4d7d-ee07-69fb49d6ca39"},"outputs":[{"name":"stdout","output_type":"stream","text":["Detection error rate = 12.4%\n"]}],"source":["from pyannote.metrics.detection import DetectionErrorRate\n","metric = DetectionErrorRate()\n","\n","for file in ami.test():\n","\n"," # apply the voice activity detection pipeline\n"," speech = pipeline(file)\n","\n"," # evaluate its output\n"," _ = metric(\n"," file['annotation'], # this is the reference annotation\n"," speech, # this is the hypothesized annotation\n"," uem=file['annotated']) # this is the part of the file that should be evaluated\n","\n","# aggregate the performance over the whole test set\n","detection_error_rate = abs(metric)\n","print(f'Detection error rate = {detection_error_rate * 100:.1f}%')"]},{"cell_type":"markdown","metadata":{"id":"CiADYH02Q-X_"},"source":["## Optimizing pipeline hyper-parameters"]},{"cell_type":"markdown","metadata":{"id":"Zik9XIQ1Q-YA"},"source":["While good enough, the hyper-parameters that we chose manually, we can try to optimize `onset` and `offset` on the development (a.k.a. validation) set to get better performance (and freeze the other two hyper-parameters)."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":219,"status":"ok","timestamp":1704893681462,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"pcQ8eJujQ-YA","outputId":"599cb846-c12f-4d6a-e5b6-c623fd9a583c"},"outputs":[],"source":["pipeline.freeze({'min_duration_on': 0.0, 'min_duration_off': 0.0})"]},{"cell_type":"code","execution_count":17,"metadata":{"executionInfo":{"elapsed":192449,"status":"ok","timestamp":1704893876383,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"xdClXgoMQ-YB"},"outputs":[],"source":["from pyannote.pipeline import Optimizer\n","\n","optimizer = Optimizer(pipeline)\n","optimizer.tune(list(ami.development()),\n"," warm_start=initial_params,\n"," n_iterations=20,\n"," show_progress=False)\n","\n","optimized_params = optimizer.best_params"]},{"cell_type":"markdown","metadata":{"id":"wVT0HjowQ-YB"},"source":["There you go: better hyper-parameters that (should) lead to better results!"]},{"cell_type":"code","execution_count":18,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":219,"status":"ok","timestamp":1704893925876,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"IuUGrd8CQ-YB","outputId":"393d66bc-b04b-4265-bbdf-19c7ff04098e"},"outputs":[{"name":"stdout","output_type":"stream","text":["{'onset': 0.824238792126195, 'offset': 0.5375254799320842, 'min_duration_on': 0.0, 'min_duration_off': 0.0}\n"]}],"source":["print(optimized_params)"]},{"cell_type":"markdown","metadata":{"id":"3nQUzBigQ-YC"},"source":["Let's evaluate the voice activity detection pipeline with this new set of hyper-parameters:"]},{"cell_type":"code","execution_count":19,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":155290,"status":"ok","timestamp":1704894084275,"user":{"displayName":"Clément PAGES","userId":"11757386314069785178"},"user_tz":-60},"id":"7A0sPO_4Q-YD","outputId":"5d28de05-fbbd-4c31-8cc7-62543c82a91d"},"outputs":[{"name":"stdout","output_type":"stream","text":["Detection error rate = 12.0%\n"]}],"source":["optimized_pipeline = pipeline.instantiate(optimized_params)\n","\n","metric = DetectionErrorRate()\n","\n","for file in ami.test():\n"," speech = optimized_pipeline(file)\n"," _ = metric(file['annotation'], speech, uem=file['annotated'])\n","\n","detection_error_rate = abs(metric)\n","print(f'Detection error rate = {detection_error_rate * 100:.1f}%')"]},{"cell_type":"markdown","metadata":{"id":"q6TiP03kQ-YD"},"source":["🎉 Yeay! We can see an improvement simply by adjusting the thresholds."]}],"metadata":{"accelerator":"GPU","colab":{"gpuType":"T4","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.10.13"},"widgets":{"application/vnd.jupyter.widget-state+json":{"0300ea5285a94c7580c789849ea8b7e7":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f3bb471baf44445d9ea50d0adde041c1","placeholder":"​","style":"IPY_MODEL_14556644c3104b02850a07f172f1411a","value":"Epoch 0: 100%"}},"14556644c3104b02850a07f172f1411a":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"146c185d279d4583860ba4f9defda025":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"290e8084737b4ff7a816f4b67fa8d411":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":"2","flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"2ea0e51154e54129a6b164fc72588edc":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"ProgressStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"321302a76845477cbfd6b644071048b9":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HBoxModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_0300ea5285a94c7580c789849ea8b7e7","IPY_MODEL_51dc1507c16e4fd88da6e9681b6f8443","IPY_MODEL_84560e1a1a154cf7a4cc998aaa7db862"],"layout":"IPY_MODEL_dce43bbd970643d99f8abbcb299925ec"}},"3b55a90beb874204a44df7f231da55eb":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_5151fae07f114d4e86ee7ec4e487faae","placeholder":"​","style":"IPY_MODEL_4d2ea6328f684e6e888bc5710dea4340","value":"Validation DataLoader 0: 100%"}},"4946ea8187904c7fa9f7c5de92a563e2":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_910983c75b5247deb203ab218e05b25a","placeholder":"​","style":"IPY_MODEL_a1861066848a4efda4cc1f09beacb451","value":" 2/2 [00:08&lt;00:00, 0.24it/s]"}},"4d2ea6328f684e6e888bc5710dea4340":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"50f8b2a4995242ac9d9961f34f2223ae":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":"2","flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"5151fae07f114d4e86ee7ec4e487faae":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"51dc1507c16e4fd88da6e9681b6f8443":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"FloatProgressModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_50f8b2a4995242ac9d9961f34f2223ae","max":136,"min":0,"orientation":"horizontal","style":"IPY_MODEL_622cf364efe34719877311f1a46d24b4","value":136}},"53ac5430d2ef45ecaaecc3da0706825b":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"622cf364efe34719877311f1a46d24b4":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"ProgressStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"69c34a612abb4d7a83bd796ef97d1c92":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":"2","flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"8429d57eec2f4c9c8cda3900c0da5df5":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"FloatProgressModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"","description":"","description_tooltip":null,"layout":"IPY_MODEL_290e8084737b4ff7a816f4b67fa8d411","max":14,"min":0,"orientation":"horizontal","style":"IPY_MODEL_bb89bf63ea124306acf411cbeaca2b90","value":14}},"84560e1a1a154cf7a4cc998aaa7db862":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_ef09269769d845fb84ba39bd43a4ee0c","placeholder":"​","style":"IPY_MODEL_b827c08a92434a31b82d0382057ee82b","value":" 136/136 [23:47&lt;00:00, 0.10it/s, v_num=0, BinaryAUROC=0.954]"}},"910983c75b5247deb203ab218e05b25a":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"929954f82bc6407aaa6b069da848d1e3":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":"inline-flex","flex":null,"flex_flow":"row wrap","grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":"hidden","width":"100%"}},"a1861066848a4efda4cc1f09beacb451":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b20cbf0c630a44f8b108d141e0b14ecc":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"FloatProgressModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"","description":"","description_tooltip":null,"layout":"IPY_MODEL_69c34a612abb4d7a83bd796ef97d1c92","max":2,"min":0,"orientation":"horizontal","style":"IPY_MODEL_2ea0e51154e54129a6b164fc72588edc","value":2}},"b62ef7013afe4da4a00ba0a70a63e938":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_53ac5430d2ef45ecaaecc3da0706825b","placeholder":"​","style":"IPY_MODEL_ffd1d20386d14e2f945f2d84df7a681f","value":"Sanity Checking DataLoader 0: 100%"}},"b827c08a92434a31b82d0382057ee82b":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b886a9d46c7647f0aa5e04260a5b6e1e":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":"inline-flex","flex":null,"flex_flow":"row wrap","grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":"hidden","width":"100%"}},"bb89bf63ea124306acf411cbeaca2b90":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"ProgressStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"d4f4fff579354b10b3cdb32cb11754fe":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e17799ed71e04167bd46363cf84cc93f","placeholder":"​","style":"IPY_MODEL_146c185d279d4583860ba4f9defda025","value":" 14/14 [01:58&lt;00:00, 0.12it/s]"}},"d5dbc90f387a4c9eb43789707505e617":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HBoxModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_3b55a90beb874204a44df7f231da55eb","IPY_MODEL_8429d57eec2f4c9c8cda3900c0da5df5","IPY_MODEL_d4f4fff579354b10b3cdb32cb11754fe"],"layout":"IPY_MODEL_b886a9d46c7647f0aa5e04260a5b6e1e"}},"dce43bbd970643d99f8abbcb299925ec":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":"inline-flex","flex":null,"flex_flow":"row wrap","grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":"100%"}},"de7bbc2d644e4ba79f3b4dcc26da3cb7":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HBoxModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_b62ef7013afe4da4a00ba0a70a63e938","IPY_MODEL_b20cbf0c630a44f8b108d141e0b14ecc","IPY_MODEL_4946ea8187904c7fa9f7c5de92a563e2"],"layout":"IPY_MODEL_929954f82bc6407aaa6b069da848d1e3"}},"e17799ed71e04167bd46363cf84cc93f":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ef09269769d845fb84ba39bd43a4ee0c":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f3bb471baf44445d9ea50d0adde041c1":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ffd1d20386d14e2f945f2d84df7a681f":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}}}}},"nbformat":4,"nbformat_minor":0}