Forums

AICovidVN 115M Challenge: Covid Cough Detection Challenge (Final Round) Forum

Go back to competition Back to thread list Post in this thread

> Đăng ký sử dụng pre-trained models (NLP / Vision / Speech)

Để công bằng giữa tất cả những người tham gia, các đội cần phải đăng ký để sử dụng các pre-trained models (NLP / Vision / Speech).
Bạn vui lòng liệt kê cụ thể các mô hình được đào tạo trước mà team sử dụng. Bạn chú ý link chung đến toàn bộ kho lưu trữ không được chấp nhận (ví dụ: link đến TorchVision sẽ không được coi là hợp lệ).

Vui lòng trả lời trong diễn đàn này để đăng ký các pre-trained models team bạn sử dụng. Với các mô hình tự train, bạn cần có link đến file ZIP, trong file Zip có (1) pre-trained weights, (2) file requirements.txt để cài môi trường, và (3) ví dụ mẫu để load pre-trained models. BTC khuyến khích các đội share pre-trained models qua github kèm link tới pre-train weights qua các dịch vụ lưu trữ file (e.g., Gdrive, Dropbox etc).

===
01. Pre-trained word embeddings: <link-to-public-repo>
===

PS. Chú ý hạn đăng ký: **14 tháng 08, 2021** - i.e., Hạn sáp nhập đội thi và đăng kí pre-trained models

Posted by: aicovidvn115m-organizers @ Aug. 5, 2021, 5:18 p.m.

Dear BTC,
Cảm ơn BTC đã tổ chức cuộc thi thú vị này.
Mình thấy quy định đăng ký pretrained models này là không hợp lý lắm, bởi vì:
- "Để công bằng giữa tất cả những người tham gia". Mình không hiểu điểm này lắm. Nếu pretrained là có sẵn trên internet thì tất cả những người tham gia đều có thể truy cập và dùng được. Vậy thì vì sao lại phải cần đăng ký pretrained?
- Hạn đăng ký pretrained là 14/8 (2 tuần trước khi kết thúc cuộc thi). Như vậy thì, tất cả các đội sẽ phải chốt models 2 tuần trước khi kết thúc cuộc thi. Và trong vòng 2 tuần cuối các đội không thể thử nghiệm model mới được nữa, điều này ảnh hưởng rất nhiều đến kết quả cuối cùng.

Đó là ý kiến cá nhân của mình. Kính mong BTC xem xét.

Posted by: none @ Aug. 6, 2021, 2:14 a.m.

Cảm ơn câu hỏi của bạn.

Mục đích của việc khai báo pre-trained model là để đảm bảo mô hình AI có thể reproduce được, điều này đặc biệt quan trọng với những cuộc thi có mục đích cộng đồng như AICovidVN. Vì vậy việc khai báo pre-trained model là cần thiết.

Với mô hình pre-trained đã được công khai, việc các thí sinh cần làm là “khai báo” - thông báo rằng mình có sử dụng mô hình này. BTC đã hướng dẫn cách thức trong bài đăng gốc.

Ngoài ra việc đăng kí mô hình pre-trained sẽ không ảnh hưởng đến cấu trúc giải pháp cuối cùng. Và các team có thể cập nhật weights cho pre-trained models, hạn cập nhật weights là trước khi bắt đầu private test.

Thân gửi,
BTC

Posted by: aicovidvn115m-organizers @ Aug. 7, 2021, 1:23 p.m.

Nếu hạn sử dụng pretrained là 14/08 thì sau thời điểm đó nếu đội thi có thay đổi kiến trúc thành mô hình khác có sử dụng pretrained khác thì cũng không thể được ạ, do đã quá hạn rồi. Nên em mong BTC xem xét lại thời gian khai báo pretrained ạ.

Posted by: nktoan @ Aug. 8, 2021, 2:48 a.m.

Chào ban tổ chức.
Mình thấy trong nội quy cuộc thi là không sử dụng bất kì dataset ngoài nào ngoại trừ data cung cấp từ cuộc thi. Vậy sao ban tổ chức lại cho phép train lại để lấy pretrained? Như thế có phải có mâu thuẫn không ạ? Mong ban tổ chức giải thích rõ phần này.

Posted by: thancaocuong @ Aug. 9, 2021, 12:56 a.m.

Chào hai bạn,

Về câu hỏi của bạn nktoan, BTC khuyến khích các đội tập trung tìm mô hình tối ưu cho pre-trained models và hoàn thành trước hạn khai báo pre-trained models. Trước hạn private, các team được phép cập nhật lại weights.

Về câu hỏi của bạn thancaocuong, BTC xin trả lời là Không mâu thuẫn bạn ạ. Đăng ký pre-trained models bao gồm đăng ký các public pre-trained models và cả các models tự pre-trained trên dữ liệu BTC cung cấp (tới thời điểm này là bao gồm dữ liệu Warm-up, dữ liệu public train, dữ liệu extra1235).

BTC

Posted by: aicovidvn115m-organizers @ Aug. 9, 2021, 9:01 a.m.

Vậy bây giờ mình lấy dữ liệu từ internet về xong tự train rồi public lên trên mạng thì nó là public model và vẫn được sử dụng trong khi mình lại sử dụng dữ liệu ngoài cuộc thi, mình hiểu vậy có phải không ?

Posted by: empty258 @ Aug. 11, 2021, 2:41 a.m.

'tf_efficientnet_b0': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth',
'tf_efficientnet_b1': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth',
'tf_efficientnet_b2': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth',
'tf_efficientnet_b3': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth',
'tf_efficientnet_b4': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth',
'tf_efficientnet_b5': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth',
'tf_efficientnet_b6': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_aa-80ba17e4.pth',
'tf_efficientnet_b7': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth',
'tf_efficientnet_b8': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth',

Posted by: none @ Aug. 11, 2021, 2:55 a.m.

Chào empty258,

Như đã nói ở trên, BTC đưa ra việc khai báo pre-trained model là để đảm bảo mô hình AI có thể reproduce được

Public pre-trained models được gọi là public khi những model này đã được đưa ra public trước khi challenge bắt đầu. Với trường hợp các mô hình nổi tiếng khác mà hữu dụng cho challenge, BTC sẽ xem xét cụ thể từng trường hợp. Ngoài ra, các mô hình khác tự train cần giới hạn trong dữ liệu của BTC cung cấp.

BTC

Posted by: aicovidvn115m-organizers @ Aug. 12, 2021, 1:32 a.m.

"Với trường hợp các mô hình nổi tiếng khác mà hữu dụng cho challenge, BTC sẽ xem xét cụ thể từng trường hợp", BTC định xem xét như thế nào, với mình khi tham gia cuộc thi thì luật là phải tuân theo, BTC xem xét thì đã làm cho cuộc thi mất tính công bằng, nếu xem xét thì phải public cách thức xem xét để công bằng cho tất cả các team

Posted by: empty258 @ Aug. 12, 2021, 3:18 a.m.

Kính gửi BTC,

Không biết là mỗi team phải đăng ký riêng pre-train hay là nếu một team đăng ký rồi thì các team khác không phải đăng ký nữa.

Team mình dùng toàn pre-train phổ biến, không biết có phải submit riêng 1 form không vậy?

Xin cảm ơn.

Posted by: VuiChoiCoThuong @ Aug. 12, 2021, 6:54 a.m.

Chào VuiChoiCoThuong,

Khi một team có sử dụng pre-trained models thì đều cần đăng ký theo như hướng dẫn ở mục FAQs: https://www.covid.aihub.vn/faqs
Việc đăng ký pre-trained ở mỗi đội là riêng biệt, không có sự phụ thuộc vào nhau.

BTC

Posted by: aicovidvn115m-organizers @ Aug. 12, 2021, 8:01 a.m.

Chào empty258,

BTC định nghĩa "Ngoại lệ" chỉ đc là các trường hợp cá biệt có thể xảy ra, nhưng quy trình đăng ký vẫn cần tuân thủ theo cách đăng ký chung. Như vậy, KHÔNG có models ngoại lệ nào mà các đội thi không được biết. BTC luôn cố gắng đảm bảo hoàn toàn tính minh bạch của các models các đội thi sử dụng.

Các đội vẫn cần đăng ký pre-trained trên forum bình thường (kể cả với các trường hợp ngoại lệ trên).

BTC

Posted by: aicovidvn115m-organizers @ Aug. 12, 2021, 8:05 a.m.

Gửi tất cả các đội,

BTC tạo kênh riêng ở thread [1] để các đội thi cùng thảo luận về việc đăng ký pre-trained models.
Các bạn vui lòng hỏi ở thread mới nếu có.

Chuyên mục này sẽ chỉ sử dụng cho việc đăng ký pre-trained models.
Các câu hỏi và các câu trả lời trước bài viết này vẫn được giữ nguyên để đảm bảo toàn vẹn thông tin.
Nhưng các post mới mà ko liên quan tới đăng ký sẽ được xoá không báo trước.

[1] https://aihub.vn/forums/22/49/

Trân trọng thông báo,
BTC

Posted by: aicovidvn115m-organizers @ Aug. 13, 2021, 12:14 a.m.

Pretrain model:
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2model
https://keras.io/api/applications/resnet/

Posted by: hungdinhvan @ Aug. 13, 2021, 8:01 a.m.

PASE: https://drive.google.com/file/d/1xwlZMGnEt9bGKCVcqDeNrruLFQW5zUEW/view
Wav2Vec: https://huggingface.co/facebook/wav2vec2-large-960h

Posted by: tinyswish @ Aug. 13, 2021, 12:38 p.m.

+) efficientnet pytorch: https://github.com/lukemelas/EfficientNet-PyTorch (efficientnet-b0, efficientnet-b1, efficientnet-b2, efficientnet-b3, efficientnet-b4, efficientnet-b5, efficientnet-b6, efficientnet-b7)
+) torchvision pretrained: resnet18/resnet34/resnet50/resnet101/resnext101_32x8d/resnext50_32x4d
+)resnest50: https://github.com/zhanghang1989/ResNeSt/releases/tag/weights_step1
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/nf_resnet50_ra2-9f236009.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/nf_regnet_b1_256_ra2-ad85cfef.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/gluon_resnest14-9c8fe254.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/gluon_resnest14-9c8fe254.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50-528c19ca.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50_fast_1s4x24d-d4a4f76f.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50_fast_4s2x40d-41d14ed0.pth
+)https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
+)https://download.pytorch.org/models/resnet18-5c106cde.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth
+) https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_v2s_ra2_288-a6477665.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mixnet_s-a907afbc.pth

Posted by: KhanhVu @ Aug. 13, 2021, 2:13 p.m.


'tf_efficientnet_bi': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_bi_aa-827b6e33.pth',
'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth',
resnet[50/18/121]+densenet121
'tf_efficientnet_b8': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth',
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2model
https://keras.io/api/applications/resnet/
https://keras.io/api/applications/densenet/
vggish:https://drive.google.com/file/d/1D2-mpFV-OSDP_dez5py79Dt80WlmVG4E/view
Yamlnet: https://www.tensorflow.org/hub/tutorials/yamnet

Posted by: meoconxinhxan @ Aug. 13, 2021, 3:27 p.m.

01. EfficientNetB4: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b5_noisy-student_notop.h5
02. EfficientNetB5: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b5_noisy-student_notop.h5
03. EfficientNetB6: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b5_noisy-student_notop.h5
04. EfficientNetB7: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b5_noisy-student_notop.h5
05. Wav2Vec2: https://github.com/huggingface/transformers

Posted by: reideen @ Aug. 13, 2021, 4:13 p.m.

resnet18: https://download.pytorch.org/models/resnet18-f37072fd.pth
resnet34: https://download.pytorch.org/models/resnet34-b627a593.pth
resnet50: https://download.pytorch.org/models/resnet50-0676ba61.pth
resnet101: https://download.pytorch.org/models/resnet101-63fe2227.pth
resnet152: https://download.pytorch.org/models/resnet152-394f9c45.pth
resnext50_32x4d: https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth
resnext101_32x8d: https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth
wide_resnet50_2: https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pth
wide_resnet101_2: https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth

Posted by: ffyytt @ Aug. 13, 2021, 6:55 p.m.

Vggish: https://github.com/tensorflow/models/tree/master/research/audioset/vggish

Posted by: thien1892 @ Aug. 14, 2021, 12:53 a.m.

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pth
https://github.com/harritaylor/torchvggish

Posted by: nhattruong @ Aug. 14, 2021, 2:32 a.m.

EfficientNetB4: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b5_noisy-student_notop.h5
EfficientNetB5: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b6_noisy-student_notop.h5
EfficientNetB6: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b7_noisy-student_notop.h5
EfficientNetB7: https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b8_noisy-student_notop.h5
Wav2Vec2: https://github.com/huggingface/transformers
Efficientnetv2 s, m, l: https://tfhub.dev/google/collections/efficientnet_v2/1

Posted by: hieuhthh @ Aug. 14, 2021, 6:01 a.m.

EfficientNetB1, EfficientNetB2, EfficientNetB3, EfficientNetB4, EfficientNetB5, ResNet18, ResNet34, ResNet50, Resnext50_32x4d: https://github.com/rwightman/pytorch-image-models
VGGish: https://github.com/harritaylor/torchvggish/releases/download/v0.1/vggish-10086976.pth
WAV2VEC: https://github.com/huggingface/transformers

Posted by: nktoan @ Aug. 14, 2021, 7:30 a.m.

https://zenodo.org/record/3987831

Posted by: tinyswish @ Aug. 14, 2021, 7:49 a.m.

'tf_efficientnet_bi': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_bi_aa-827b6e33.pth',
'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth',
resnet[50/18/121]+densenet121
'tf_efficientnet_b8': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth',
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2model
https://keras.io/api/applications/resnet/
https://keras.io/api/applications/densenet/
vggish:https://drive.google.com/file/d/1D2-mpFV-OSDP_dez5py79Dt80WlmVG4E/view
Yamlnet: https://www.tensorflow.org/hub/tutorials/yamnet
PANN *: https://zenodo.org/record/3987831#.YRd7OlsxWo5
'densenet121': url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth'),
'densenetblur121d': url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenetblur121d_ra-100dcfbc.pth'),
'densenet169': _cfg(url='https://download.pytorch.org/models/densenet169-b2777c0a.pth'
'densenet201': _cfg(url='https://download.pytorch.org/models/densenet201-c1103571.pth'
'densenet161': _cfg(url='https://download.pytorch.org/models/densenet161-8d451a50.pth'
'densenet264': _cfg(url=''),
'densenet264d_iabn': _cfg(url=''),
'tv_densenet121': _cfg(url='https://download.pytorch.org/models/densenet121-a639ec97.pth'
'resnet18': _cfg(url='https://download.pytorch.org/models/resnet18-5c106cde.pth'),
'resnet18d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth',
interpolation='bicubic', first_conv='conv1.0'),
'resnet34': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth'),
'resnet34d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth',
interpolation='bicubic', first_conv='conv1.0'),
'resnet26': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26-9aa10e23.pth',
interpolation='bicubic'),
'resnet26d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth',
interpolation='bicubic', first_conv='conv1.0'),
'resnet26t': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8)),
'resnet50': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth',
interpolation='bicubic'),
'resnet50d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth',
interpolation='bicubic', first_conv='conv1.0'),
'resnet50t': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0'),
'resnet101': _cfg(url='', interpolation='bicubic'),
'resnet101d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet101d_ra2-2803ffab.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=1.0, test_input_size=(3, 320, 320)),
'resnet152': _cfg(url='', interpolation='bicubic'),
'resnet152d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet152d_ra2-5cac0439.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=1.0, test_input_size=(3, 320, 320)),
'resnet200': _cfg(url='', interpolation='bicubic'),
'resnet200d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet200d_ra2-bdba9bf9.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=1.0, test_input_size=(3, 320, 320)),
'tv_resnet34': _cfg(url='https://download.pytorch.org/models/resnet34-333f7ec4.pth'),
'tv_resnet50': _cfg(url='https://download.pytorch.org/models/resnet50-19c8e357.pth'),
'tv_resnet101': _cfg(url='https://download.pytorch.org/models/resnet101-5d3b4d8f.pth'),
'tv_resnet152': _cfg(url='https://download.pytorch.org/models/resnet152-b121ed2d.pth'),
'wide_resnet50_2': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/wide_resnet50_racm-8234f177.pth',
interpolation='bicubic'),
'wide_resnet101_2': _cfg(url='https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth'),

# ResNeXt
'resnext50_32x4d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth',
interpolation='bicubic'),
'resnext50d_32x4d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50d_32x4d-103e99f8.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'resnext101_32x4d': _cfg(url=''),
'resnext101_32x8d': _cfg(url='https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth'),
'resnext101_64x4d': _cfg(url=''),
'tv_resnext50_32x4d': _cfg(url='https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth'),

# ResNeXt models - Weakly Supervised Pretraining on Instagram Hashtags
# from https://github.com/facebookresearch/WSL-Images
# Please note the CC-BY-NC 4.0 license on theses weights, non-commercial use only.
'ig_resnext101_32x8d': _cfg(url='https://download.pytorch.org/models/ig_resnext101_32x8-c38310e5.pth'),
'ig_resnext101_32x16d': _cfg(url='https://download.pytorch.org/models/ig_resnext101_32x16-c6f796b0.pth'),
'ig_resnext101_32x32d': _cfg(url='https://download.pytorch.org/models/ig_resnext101_32x32-e4b90b00.pth'),
'ig_resnext101_32x48d': _cfg(url='https://download.pytorch.org/models/ig_resnext101_32x48-3e41cc8a.pth'),

# Semi-Supervised ResNe*t models from https://github.com/facebookresearch/semi-supervised-ImageNet1K-models
# Please note the CC-BY-NC 4.0 license on theses weights, non-commercial use only.
'ssl_resnet18': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet18-d92f0530.pth'),
'ssl_resnet50': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet50-08389792.pth'),
'ssl_resnext50_32x4d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext50_32x4-ddb3e555.pth'),
'ssl_resnext101_32x4d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x4-dc43570a.pth'),
'ssl_resnext101_32x8d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x8-2cfe2f8b.pth'),
'ssl_resnext101_32x16d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x16-15fffa57.pth'),

# Semi-Weakly Supervised ResNe*t models from https://github.com/facebookresearch/semi-supervised-ImageNet1K-models
# Please note the CC-BY-NC 4.0 license on theses weights, non-commercial use only.
'swsl_resnet18': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnet18-118f1556.pth'),
'swsl_resnet50': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnet50-16a12f1b.pth'),
'swsl_resnext50_32x4d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext50_32x4-72679e44.pth'),
'swsl_resnext101_32x4d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x4-3f87e46b.pth'),
'swsl_resnext101_32x8d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x8-b4712904.pth'),
'swsl_resnext101_32x16d': _cfg(
url='https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x16-f3559a9c.pth'),

# Squeeze-Excitation ResNets, to eventually replace the models in senet.py
'seresnet18': _cfg(
url='',
interpolation='bicubic'),
'seresnet34': _cfg(
url='',
interpolation='bicubic'),
'seresnet50': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet50_ra_224-8efdb4bb.pth',
interpolation='bicubic'),
'seresnet50t': _cfg(
url='',
interpolation='bicubic',
first_conv='conv1.0'),
'seresnet101': _cfg(
url='',
interpolation='bicubic'),
'seresnet152': _cfg(
url='',
interpolation='bicubic'),
'seresnet152d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet152d_ra2-04464dd2.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=1.0, test_input_size=(3, 320, 320)
),
'seresnet200d': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), crop_pct=0.94, pool_size=(8, 8)),
'seresnet269d': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), crop_pct=0.94, pool_size=(8, 8)),

# Squeeze-Excitation ResNeXts, to eventually replace the models in senet.py
'seresnext26d_32x4d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26d_32x4d-80fa48a3.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'seresnext26t_32x4d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26tn_32x4d-569cb627.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'seresnext50_32x4d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext50_32x4d_racm-a304a460.pth',
interpolation='bicubic'),
'seresnext101_32x4d': _cfg(
url='',
interpolation='bicubic'),
'seresnext101_32x8d': _cfg(
url='',
interpolation='bicubic'),
'senet154': _cfg(
url='',
interpolation='bicubic',
first_conv='conv1.0'),

# Efficient Channel Attention ResNets
'ecaresnet26t': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet26t_ra2-46609757.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=0.95, test_input_size=(3, 320, 320)),
'ecaresnetlight': _cfg(
url='https://imvl-automl-sh.oss-cn-shanghai.aliyuncs.com/darts/hyperml/hyperml/job_45402/outputs/ECAResNetLight_4f34b35b.pth',
interpolation='bicubic'),
'ecaresnet50d': _cfg(
url='https://imvl-automl-sh.oss-cn-shanghai.aliyuncs.com/darts/hyperml/hyperml/job_45402/outputs/ECAResNet50D_833caf58.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'ecaresnet50d_pruned': _cfg(
url='https://imvl-automl-sh.oss-cn-shanghai.aliyuncs.com/darts/hyperml/hyperml/job_45899/outputs/ECAResNet50D_P_9c67f710.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'ecaresnet50t': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet50t_ra2-f7ac63c4.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), pool_size=(8, 8),
crop_pct=0.95, test_input_size=(3, 320, 320)),
'ecaresnet101d': _cfg(
url='https://imvl-automl-sh.oss-cn-shanghai.aliyuncs.com/darts/hyperml/hyperml/job_45402/outputs/ECAResNet101D_281c5844.pth',
interpolation='bicubic', first_conv='conv1.0'),
'ecaresnet101d_pruned': _cfg(
url='https://imvl-automl-sh.oss-cn-shanghai.aliyuncs.com/darts/hyperml/hyperml/job_45610/outputs/ECAResNet101D_P_75a3370e.pth',
interpolation='bicubic',
first_conv='conv1.0'),
'ecaresnet200d': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 256, 256), crop_pct=0.94, pool_size=(8, 8)),
'ecaresnet269d': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet269d_320_ra2-7baa55cb.pth',
interpolation='bicubic', first_conv='conv1.0', input_size=(3, 320, 320), pool_size=(10, 10),
crop_pct=1.0, test_input_size=(3, 352, 352)),

# Efficient Channel Attention ResNeXts
'ecaresnext26t_32x4d': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0'),
'ecaresnext50t_32x4d': _cfg(
url='',
interpolation='bicubic', first_conv='conv1.0'),

# ResNets with anti-aliasing blur pool
'resnetblur18': _cfg(
interpolation='bicubic'),
'resnetblur50': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnetblur50-84f4748f.pth',
interpolation='bicubic'),

# ResNet-RS models
'resnetrs50': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs50_ema-6b53758b.pth',
input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.91, test_input_size=(3, 224, 224),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs101': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs101_i192_ema-1509bbf6.pth',
input_size=(3, 192, 192), pool_size=(6, 6), crop_pct=0.94, test_input_size=(3, 288, 288),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs152': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs152_i256_ema-a9aff7f9.pth',
input_size=(3, 256, 256), pool_size=(8, 8), crop_pct=1.0, test_input_size=(3, 320, 320),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs200': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs200_ema-623d2f59.pth',
input_size=(3, 256, 256), pool_size=(8, 8), crop_pct=1.0, test_input_size=(3, 320, 320),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs270': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs270_ema-b40e674c.pth',
input_size=(3, 256, 256), pool_size=(8, 8), crop_pct=1.0, test_input_size=(3, 352, 352),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs350': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs350_i256_ema-5a1aa8f1.pth',
input_size=(3, 288, 288), pool_size=(9, 9), crop_pct=1.0, test_input_size=(3, 384, 384),
interpolation='bicubic', first_conv='conv1.0'),
'resnetrs420': _cfg(
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs420_ema-972dee69.pth',
input_size=(3, 320, 320), pool_size=(10, 10), crop_pct=1.0, test_input_size=(3, 416, 416),
interpolation='bicubic', first_conv='conv1.0'),

Posted by: meoconxinhxan @ Aug. 14, 2021, 8:17 a.m.

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_ns-c0e6a31c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_ns-99dd0c41.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_ns-00306e48.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_ns-9d44bf68.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_ns-d6313a46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ns-6f26d0cf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_ns-51548356.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ns-1dbc32de.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth

Posted by: gengar @ Aug. 14, 2021, 8:27 a.m.

Pretrain model we are considering to use:
https://huggingface.co/facebook/wav2vec2-base-960h/blob/main/tf_model.h5
https://huggingface.co/facebook/wav2vec2-large-960h/blob/main/pytorch_model.bin
https://storage.googleapis.com/keras-applications/efficientnetb0.h5
https://storage.googleapis.com/keras-applications/efficientnetb1.h5
https://storage.googleapis.com/keras-applications/efficientnetb2.h5
https://storage.googleapis.com/keras-applications/efficientnetb3.h5
https://storage.googleapis.com/keras-applications/efficientnetb4.h5
https://storage.googleapis.com/tensorflow/keras-applications/inception_resnet_v2/inception_resnet_v2_weights_tf_dim_ordering_tf_kernels.h5
https://storage.googleapis.com/keras-applications/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5
https://pytorch.org/hub/snakers4_silero-vad_vad/
https://storage.googleapis.com/keras-applications/resnet/resnet50v2_weights_tf_dim_ordering_tf_kernels_notop
https://storage.googleapis.com/tensorflow/keras-applications/densenet/densenet121_weights_tf_dim_ordering_tf_kernels_notop.h5
https://storage.googleapis.com/tensorflow/keras-applications/densenet/densenet169_weights_tf_dim_ordering_tf_kernels_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet34_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/wide_resnet101_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/wide_resnet50_imagenet_notop.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet18_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet34_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet50_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnext50_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnext50_imagenet_1000_no_top_v2.h5.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/resnet50_imagenet11k-places365ch_11586_no_top.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b0-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b1-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b2-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b3-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-s-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-m-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-l-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b0-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b1-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b2-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b3-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-s-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-m-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-l-21k-ft1k.h5
https://tfhub.dev/google/vggish/1
https://tfhub.dev/google/yamnet/1
https://drive.google.com/file/d/12Kv9S7E3DuRjoFIIPPmlmfELjEV0r-q8/view?usp=sharing
https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b0_noisy-student_notop.h5
https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b1_noisy-student_notop.h5
https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b2_noisy-student_notop.h5
https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b3_noisy-student_notop.h5

Posted by: ZALO_NGUYEN_QUAN_ANH_MINH @ Aug. 14, 2021, 9:18 a.m.

https://github.com/tensorflow/models/tree/master/research/audioset
https://keras.io/api/applications/resnet/
https://keras.io/api/applications/densenet/

Posted by: vokhanhan25 @ Aug. 14, 2021, 10:09 a.m.

https://keras.io/api/applications/resnet/#resnet50-function
https://keras.io/api/applications/resnet/#resnet101-function
https://keras.io/api/applications/resnet/#resnet152-function
https://keras.io/api/applications/resnet/#resnet50v2-function
https://keras.io/api/applications/resnet/#resnet101v2-function
https://keras.io/api/applications/resnet/#resnet152v2-function
https://keras.io/api/applications/inceptionv3
https://keras.io/api/applications/inceptionresnetv2
https://keras.io/api/applications/densenet/#densenet121-function
https://keras.io/api/applications/densenet/#densenet169-function
https://keras.io/api/applications/densenet/#densenet201-function
https://www.tensorflow.org/hub/tutorials/yamnet
https://github.com/tensorflow/models/tree/master/research/audioset/vggish

Posted by: dragon_vn @ Aug. 14, 2021, 11:33 a.m.

1. https://keras.io/api/applications/densenet/
2. https://keras.io/api/applications/resnet

Posted by: SauSun @ Aug. 14, 2021, 12:15 p.m.

https://github.com/tensorflow/models/tree/master/research/audioset
https://keras.io/api/applications/resnet/
https://keras.io/api/applications/densenet/
https://keras.io/api/applications/efficientnet

Posted by: aianticov2 @ Aug. 14, 2021, 12:30 p.m.

https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/2

Posted by: vhnguyen @ Aug. 14, 2021, 12:33 p.m.

"https://keras.io/api/applications/resnet/"
"https://keras.io/api/applications/densenet/
#xception:
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-cadene/xception-43020ad28.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_xception_41-e6439c97.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_xception_65-c9ae96e8.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_xception_71-8eec7df1.pth"
#efficientnet
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_aa-80ba17e4.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b0-c7cc451f.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b1-be6e41b0.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b2-847de54e.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b3-57773f13.pth"
#pann
"https://zenodo.org/record/3987831#.YRd7OlsxWo5"
"https://zenodo.org/record/3987831/files/Cnn10_mAP%3D0.380.pth"
"https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth"
"https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth"
"https://zenodo.org/record/3987831/files/Cnn14_DecisionLevelAtt_mAP%3D0.425.pth"
"https://zenodo.org/record/3987831/files/ResNet22_mAP%3D0.430.pth"
"https://zenodo.org/record/3987831/files/ResNet38_mAP%3D0.434.pth"
"https://zenodo.org/record/3987831/files/ResNet54_mAP%3D0.429.pth"
"https://zenodo.org/record/3987831/files/Wavegram_Logmel_Cnn14_mAP%3D0.439.pth"
#inception
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_inception_v3-e0069de4.pth"
#own_pretrained

#densenet
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenetblur121d_ra-100dcfbc.pth"
"https://download.pytorch.org/models/densenet201-c1103571.pth"
"https://download.pytorch.org/models/densenet161-8d451a50.pth"
#resnet
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet101d_ra2-2803ffab.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth"

Posted by: wiseking @ Aug. 14, 2021, 1:02 p.m.

Resnet18: https://download.pytorch.org/models/resnet18-f37072fd.pth
Resnet34: https://download.pytorch.org/models/resnet34-b627a593.pth
Resnet50: https://download.pytorch.org/models/resnet50-0676ba61.pth
Resnet101: https://download.pytorch.org/models/resnet101-63fe2227.pth
Resnet152: https://download.pytorch.org/models/resnet152-394f9c45.pth
Resnext50_32x4d: https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth
Resnext: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth
Resnext: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50d_32x4d-103e99f8.pth
Resnext101_32x8d: https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth
Efficientnet : https://github.com/lukemelas/EfficientNet-PyTorch (efficientnet-b0, efficientnet-b1, efficientnet-b2, efficientnet-b3, efficientnet-b4, efficientnet-b5, efficientnet-b6, efficientnet-b7)
Densenet121: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth
Densenet161: https://download.pytorch.org/models/densenet161-8d451a50.pth
Densenet169: https://download.pytorch.org/models/densenet169-b2777c0a.pth
Densenet201: https://download.pytorch.org/models/densenet201-c1103571.pth
Wav2Vec2: https://github.com/huggingface/transformers

Posted by: hhoanguet @ Aug. 14, 2021, 1:27 p.m.

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_ns-c0e6a31c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_ns-99dd0c41.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_ns-00306e48.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_ns-9d44bf68.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_ns-d6313a46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ns-6f26d0cf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_ns-51548356.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ns-1dbc32de.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth
'Wav2Vec2': 'https://github.com/huggingface/transformers'
https://huggingface.co/facebook/wav2vec2-base-960h/blob/main/tf_model.h5
https://huggingface.co/facebook/wav2vec2-large-960h/blob/main/pytorch_model.bin
'resnet': 'https://keras.io/api/applications/resnet/'
'densenet': 'https://keras.io/api/applications/densenet/'
'efficientnet': 'https://keras.io/api/applications/efficientnet'
'yamnet': 'https://tfhub.dev/google/yamnet/1'
'vggish': 'https://storage.googleapis.com/audioset/vggish_model.ckpt'
'TRILL': 'https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/3'
https://storage.googleapis.com/keras-applications/efficientnetb0.h5
https://storage.googleapis.com/keras-applications/efficientnetb1.h5
https://storage.googleapis.com/keras-applications/efficientnetb2.h5
https://storage.googleapis.com/keras-applications/efficientnetb3.h5
https://storage.googleapis.com/keras-applications/efficientnetb4.h5
https://storage.googleapis.com/tensorflow/keras-applications/inception_resnet_v2/inception_resnet_v2_weights_tf_dim_ordering_tf_kernels.h5
https://storage.googleapis.com/keras-applications/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5
https://storage.googleapis.com/keras-applications/resnet/resnet50v2_weights_tf_dim_ordering_tf_kernels_notop
https://storage.googleapis.com/tensorflow/keras-applications/densenet/densenet121_weights_tf_dim_ordering_tf_kernels_notop.h5
https://storage.googleapis.com/tensorflow/keras-applications/densenet/densenet169_weights_tf_dim_ordering_tf_kernels_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet18_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet34_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.3.0/resnet50_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext101_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_ssl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/resnext50_swsl_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/wide_resnet101_imagenet_notop.h5
https://github.com/RichardXiao13/TensorFlow-ResNets/releases/download/v0.1.0/wide_resnet50_imagenet_notop.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet18_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet34_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnet50_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnext50_imagenet_1000_no_top.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/seresnext50_imagenet_1000_no_top_v2.h5.h5
https://github.com/qubvel/classification_models/releases/download/0.0.1/resnet50_imagenet11k-places365ch_11586_no_top.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b0-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b1-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b2-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b3-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-s-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-m-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-l-21k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b0-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b1-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b2-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-b3-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-s-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-m-21k-ft1k.h5
https://github.com/leondgarse/keras_efficientnet_v2/releases/download/v1.0.0/efficientnetv2-l-21k-ft1k.h5

Posted by: WhyNot @ Aug. 14, 2021, 1:47 p.m.

https://download.pytorch.org/models/vgg19-dcbb9e9d.pth
https://download.pytorch.org/models/vgg16-397923af.pth

https://download.pytorch.org/models/resnet18-f37072fd.pth
https://download.pytorch.org/models/resnet34-b627a593.pth
https://download.pytorch.org/models/resnet50-0676ba61.pth
https://download.pytorch.org/models/resnet101-63fe2227.pth
https://download.pytorch.org/models/resnet152-394f9c45.pth
https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth
https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth
https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pth
https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth

https://download.pytorch.org/models/densenet121-a639ec97.pth
https://download.pytorch.org/models/densenet169-b2777c0a.pth
https://download.pytorch.org/models/densenet201-c1103571.pth
https://download.pytorch.org/models/densenet161-8d451a50.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet50_ra_224-8efdb4bb.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet152d_ra2-04464dd2.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26d_32x4d-80fa48a3.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26tn_32x4d-569cb627.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext50_32x4d_racm-a304a460.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50d_32x4d-103e99f8.pth

https://storage.googleapis.com/vit_models/augreg/Ti_16-i21k-300ep-lr_0.001-aug_none-wd_0.03-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_224.npz
https://storage.googleapis.com/vit_models/augreg/S_32-i21k-300ep-lr_0.001-aug_light1-wd_0.03-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_224.npz
https://storage.googleapis.com/vit_models/augreg/S_32-i21k-300ep-lr_0.001-aug_light1-wd_0.03-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_384.npz
https://storage.googleapis.com/vit_models/augreg/S_16-i21k-300ep-lr_0.001-aug_light1-wd_0.03-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_224.npz
https://storage.googleapis.com/vit_models/augreg/S_16-i21k-300ep-lr_0.001-aug_light1-wd_0.03-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_384.npz
https://storage.googleapis.com/vit_models/augreg/B_32-i21k-300ep-lr_0.001-aug_light1-wd_0.1-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.03-res_384.npz
https://storage.googleapis.com/vit_models/augreg/B_16-i21k-300ep-lr_0.001-aug_medium1-wd_0.1-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.01-res_384.npz
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-vitjx/jx_vit_large_p32_384-9b920ba8.pth
https://storage.googleapis.com/vit_models/augreg/L_16-i21k-300ep-lr_0.001-aug_medium1-wd_0.1-do_0.1-sd_0.1--imagenet2012-steps_20k-lr_0.01-res_384.npz
https://storage.googleapis.com/vit_models/augreg/S_32-i21k-300ep-lr_0.001-aug_light1-wd_0.03-do_0.0-sd_0.0.npz

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b0_ra-3dd342df.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b1-533bc792.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b2_ra-bcdf34b7.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b3_ra2-cf984f9c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b4_ra2_320-7eb33cd5.pth

https://download.pytorch.org/models/inception_v3_google-1a9a5a14.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_inception_v3-e0069de4.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv3_large_100_ra-f55367f5.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_mobilenetv3_large_075-150ee8b0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_mobilenetv3_large_100-427764d5.pth

Posted by: BUIBAANH @ Aug. 14, 2021, 2:02 p.m.

Team Phoenix khai báo các pretrained models:
1. AST: Audio Spectrogram Transformer (https://github.com/YuanGongND/ast)
Pre 1: https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
Pre 2: https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
Pre 3: https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
Pre 4: https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
Pre 5: https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
Pre 6: https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
Pre 7: https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

2. PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition (https://github.com/qiuqiangkong/audioset_tagging_cnn)
Pre: https://zenodo.org/record/3987831#.YRfIAIgzaK0 (toàn bộ phần Files)

3. VGGish (https://github.com/harritaylor/torchvggish)
Pre1: https://storage.googleapis.com/audioset/vggish_model.ckpt
Params: https://storage.googleapis.com/audioset/vggish_pca_params.npz

4. Timm (https://github.com/rwightman/pytorch-image-models)
Pre: Tất cả các default pretrained của timm, phiên bản mới nhất.

5. Yamnet (https://github.com/tensorflow/models/tree/master/research/audioset/yamnet , https://github.com/w-hc/torch_audioset)
Pre: https://storage.googleapis.com/audioset/yamnet.h5

6. Wav2Vec (https://huggingface.co/transformers/model_doc/wav2vec2.html)
Pre: https://github.com/pytorch/fairseq/blob/master/examples/wav2vec/README.md

7. Silero VAD (https://github.com/snakers4/silero-vad#pre-trained-models)
Pre: Các link colab trong readme.

Posted by: yoongaehwa @ Aug. 14, 2021, 2:06 p.m.

Bổ sung:
"https://zenodo.org/record/3987831#.YRd7OlsxWo5"
"https://zenodo.org/record/3987831/files/Cnn10_mAP%3D0.380.pth"
"https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth"
"https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth"
"https://zenodo.org/record/3987831/files/Cnn14_DecisionLevelAtt_mAP%3D0.425.pth"
"https://zenodo.org/record/3987831/files/ResNet22_mAP%3D0.430.pth"
"https://zenodo.org/record/3987831/files/ResNet38_mAP%3D0.434.pth"
"https://zenodo.org/record/3987831/files/ResNet54_mAP%3D0.429.pth"
"https://zenodo.org/record/3987831/files/Wavegram_Logmel_Cnn14_mAP%3D0.439.pth"
#inception
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_inception_v3-e0069de4.pth"
#own_pretrained

#densenet
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenetblur121d_ra-100dcfbc.pth"
"https://download.pytorch.org/models/densenet201-c1103571.pth"
"https://download.pytorch.org/models/densenet161-8d451a50.pth"

Posted by: KhanhVu @ Aug. 14, 2021, 2:22 p.m.

Bổ sung:
https://download.pytorch.org/models/densenet121-a639ec97.pth
https://download.pytorch.org/models/densenet169-b2777c0a.pth
https://download.pytorch.org/models/densenet201-c1103571.pth
https://download.pytorch.org/models/densenet161-8d451a50.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet50_ra_224-8efdb4bb.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet152d_ra2-04464dd2.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26d_32x4d-80fa48a3.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26tn_32x4d-569cb627.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext50_32x4d_racm-a304a460.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50d_32x4d-103e99f8.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth
VGGish: https://github.com/harritaylor/torchvggish/releases/download/v0.1/vggish-10086976.pth
WAV2VEC: https://github.com/huggingface/transformers
https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pth

Posted by: KhanhVu @ Aug. 14, 2021, 2:36 p.m.

"https://zenodo.org/record/3987831#.YRd7OlsxWo5"
"https://zenodo.org/record/3987831/files/Cnn10_mAP%3D0.380.pth"
"https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth"
"https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth"
"https://zenodo.org/record/3987831/files/Cnn14_DecisionLevelAtt_mAP%3D0.425.pth"
"https://zenodo.org/record/3987831/files/ResNet22_mAP%3D0.430.pth"
"https://zenodo.org/record/3987831/files/ResNet38_mAP%3D0.434.pth"
"https://zenodo.org/record/3987831/files/ResNet54_mAP%3D0.429.pth"
"https://zenodo.org/record/3987831/files/Wavegram_Logmel_Cnn14_mAP%3D0.439.pth
'tf_efficientnet_b0': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth',
'tf_efficientnet_b1': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth',
'tf_efficientnet_b2': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth',
'tf_efficientnet_b3': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth',
'tf_efficientnet_b4': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth',
'tf_efficientnet_b5': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth',
'tf_efficientnet_b6': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_aa-80ba17e4.pth',
'tf_efficientnet_b7': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth',
'tf_efficientnet_b8': 'https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth',
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2model
https://audeering.github.io/opensmile/get-started.html#extracting-your-first-features
https://keras.io/api/applications/resnet/
https://keras.io/api/applications/densenet/
vggish:https://drive.google.com/file/d/1D2-mpFV-OSDP_dez5py79Dt80WlmVG4E/view
Yamlnet: https://www.tensorflow.org/hub/tutorials/yamnet
VGGish: https://github.com/harritaylor/torchvggish/releases/download/v0.1/vggish-10086976.pth
WAV2VEC: https://github.com/huggingface/transformers
https://tfhub.dev/google/yamnet/1?tf-hub-format=compressed
https://drive.google.com/file/d/1D2-mpFV-OSDP_dez5py79Dt80WlmVG4E/view
https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/2
'yamnet': 'https://tfhub.dev/google/yamnet/1'
'vggish': 'https://storage.googleapis.com/audioset/vggish_model.ckpt'
https://www.tensorflow.org/hub/tutorials/yamnet
https://github.com/tensorflow/models/tree/master/research/audioset/vggish
https://huggingface.co/facebook/wav2vec2-base-960h/blob/main/tf_model.h5
https://huggingface.co/facebook/wav2vec2-large-960h/blob/main/pytorch_model.bin
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_ns-c0e6a31c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_ns-99dd0c41.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_ns-00306e48.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_ns-9d44bf68.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_ns-d6313a46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ns-6f26d0cf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_ns-51548356.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ns-1dbc32de.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth
1. AST: Audio Spectrogram Transformer (https://github.com/YuanGongND/ast)
Pre 1: https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
Pre 2: https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
Pre 3: https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
Pre 4: https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
Pre 5: https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
Pre 6: https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
Pre 7: https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

2. PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition (https://github.com/qiuqiangkong/audioset_tagging_cnn)
Pre: https://zenodo.org/record/3987831#.YRfIAIgzaK0 (toàn bộ phần Files)

3. VGGish (https://github.com/harritaylor/torchvggish)
Pre1: https://storage.googleapis.com/audioset/vggish_model.ckpt
Params: https://storage.googleapis.com/audioset/vggish_pca_params.npz

4. Timm (https://github.com/rwightman/pytorch-image-models)
Pre: Tất cả các default pretrained của timm, phiên bản mới nhất.

5. Yamnet (https://github.com/tensorflow/models/tree/master/research/audioset/yamnet , https://github.com/w-hc/torch_audioset)
Pre: https://storage.googleapis.com/audioset/yamnet.h5

6. Wav2Vec (https://huggingface.co/transformers/model_doc/wav2vec2.html)
Pre: https://github.com/pytorch/fairseq/blob/master/examples/wav2vec/README.md

7. Silero VAD (https://github.com/snakers4/silero-vad#pre-trained-models)
Pre: Các link colab trong readme.

Posted by: none @ Aug. 14, 2021, 2:40 p.m.

AST: Audio Spectrogram Transformer (https://github.com/YuanGongND/ast)
Pre 1: https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
Pre 2: https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
Pre 3: https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
Pre 4: https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
Pre 5: https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
Pre 6: https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
Pre 7: https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

Posted by: KhanhVu @ Aug. 14, 2021, 2:46 p.m.

Bổ sung:
VGGish: https://github.com/harritaylor/torchvggish/releases/download/v0.1/vggish-10086976.pth
Wav2Vec: https://github.com/huggingface/transformers
Panns : https://zenodo.org/record/3987831#.YRfbe4gplPa (all file)
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_aa-80ba17e4.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ra-6c08e654.pth"
"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b8_ra-572d5dd9.pth"

Posted by: BUIBAANH @ Aug. 14, 2021, 3:06 p.m.

VGGish: https://github.com/harritaylor/torchvggish/releases/download/v0.1/vggish-10086976.pth
Inception: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_inception_v3-e0069de4.pth
https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth?download=1
https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth?download=1
https://zenodo.org/record/3987831/files/Wavegram_Logmel_Cnn14_mAP%3D0.439.pth?download=1
https://zenodo.org/record/3987831/files/ResNet54_mAP%3D0.429.pth?download=1
https://zenodo.org/record/3987831/files/ResNet38_mAP%3D0.434.pth?download=1
https://zenodo.org/record/3987831/files/ResNet22_mAP%3D0.430.pth?download=1
https://zenodo.org/record/3987831/files/Cnn10_mAP%3D0.380.pth
https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth
https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth
https://zenodo.org/record/3987831/files/Cnn14_DecisionLevelAtt_mAP%3D0.425.pth
Audio Spectrogram Transformer (https://github.com/YuanGongND/ast)
Pre 1: https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
Pre 2: https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
Pre 3: https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
Pre 4: https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
Pre 5: https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
Pre 6: https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
Pre 7: https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1
Yamnet (https://github.com/tensorflow/models/tree/master/research/audioset/yamnet , https://github.com/w-hc/torch_audioset)
Pre: https://storage.googleapis.com/audioset/yamnet.h5

Posted by: hhoanguet @ Aug. 14, 2021, 3:11 p.m.

https://keras.io/api/applications/xception
https://keras.io/api/applications/vgg/#vgg16-function
https://keras.io/api/applications/vgg/#vgg19-function
https://keras.io/api/applications/resnet/#resnet50-function
https://keras.io/api/applications/resnet/#resnet101-function
https://keras.io/api/applications/resnet/#resnet152-function
https://keras.io/api/applications/resnet/#resnet50v2-function
https://keras.io/api/applications/resnet/#resnet101v2-function
https://keras.io/api/applications/resnet/#resnet152v2-function
https://keras.io/api/applications/inceptionv3
https://keras.io/api/applications/inceptionresnetv2
https://keras.io/api/applications/mobilenet
https://keras.io/api/applications/mobilenet/#mobilenetv2-function
https://keras.io/api/applications/densenet/#densenet121-function
https://keras.io/api/applications/densenet/#densenet169-function
https://keras.io/api/applications/densenet/#densenet201-function
https://keras.io/api/applications/nasnet/#nasnetmobile-function
https://keras.io/api/applications/nasnet/#nasnetlarge-function
https://keras.io/api/applications/efficientnet/#efficientnetb0-function
https://keras.io/api/applications/efficientnet/#efficientnetb1-function
https://keras.io/api/applications/efficientnet/#efficientnetb2-function
https://keras.io/api/applications/efficientnet/#efficientnetb3-function
https://keras.io/api/applications/efficientnet/#efficientnetb4-function
https://keras.io/api/applications/efficientnet/#efficientnetb5-function
https://keras.io/api/applications/efficientnet/#efficientnetb6-function
https://keras.io/api/applications/efficientnet/#efficientnetb7-function
https://www.tensorflow.org/hub/tutorials/yamnet
https://github.com/tensorflow/models/tree/master/research/audioset/vggish
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2model
https://huggingface.co/transformers/model_doc/wav2vec2.html#tfwav2vec2forctc
https://huggingface.co/facebook/wav2vec2-base-960h
https://huggingface.co/facebook/wav2vec2-large-960h
https://pytorch.org/hub/pytorch_vision_resnet/
https://pytorch.org/hub/pytorch_vision_resnext/
https://pytorch.org/hub/pytorch_vision_fcn_resnet101/
https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/2
https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/3
https://zenodo.org/record/3987831#.YRfU1zczZhE
https://zenodo.org/record/3987831#.YRd7OlsxWo5
https://zenodo.org/record/3987831/files/Cnn10_mAP%3D0.380.pth
https://zenodo.org/record/3987831/files/Cnn14_16k_mAP%3D0.438.pth
https://zenodo.org/record/3987831/files/Cnn14_8k_mAP%3D0.416.pth
https://zenodo.org/record/3987831/files/Cnn14_DecisionLevelAtt_mAP%3D0.425.pth
https://zenodo.org/record/3987831/files/ResNet22_mAP%3D0.430.pth
https://zenodo.org/record/3987831/files/ResNet38_mAP%3D0.434.pth
https://zenodo.org/record/3987831/files/ResNet54_mAP%3D0.429.pth
https://zenodo.org/record/3987831/files/Wavegram_Logmel_Cnn14_mAP%3D0.439.pth
https://huggingface.co/facebook/wav2vec2-base-960h/blob/main/tf_model.h5
https://huggingface.co/facebook/wav2vec2-large-960h/blob/main/pytorch_model.bin
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_ns-c0e6a31c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_ns-99dd0c41.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_ns-00306e48.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_ns-9d44bf68.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_ns-d6313a46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ns-6f26d0cf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b6_ns-51548356.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b7_ns-1dbc32de.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b4_aa-818f208c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b5_ra-9a3e5369.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/gluon_resnest14-9c8fe254.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/gluon_resnest26-50eb607c.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50-528c19ca.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest101-22405ba7.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest200-75117900.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest269-0cc87c48.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50_fast_4s2x40d-41d14ed0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-resnest/resnest50_fast_1s4x24d-d4a4f76f.pth
https://storage.googleapis.com/bit_models/BiT-M-R50x1-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R50x3-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R101x1-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R101x3-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R152x2-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R152x4-ILSVRC2012.npz
https://storage.googleapis.com/bit_models/BiT-M-R50x1.npz
https://storage.googleapis.com/bit_models/BiT-M-R50x3.npz
https://storage.googleapis.com/bit_models/BiT-M-R101x1.npz
https://storage.googleapis.com/bit_models/BiT-M-R101x3.npz
https://storage.googleapis.com/bit_models/BiT-M-R152x2.npz
https://storage.googleapis.com/bit_models/BiT-M-R152x4.npz
https://storage.googleapis.com/bit_models/distill/R50x1_224.npz
https://storage.googleapis.com/bit_models/distill/R152x2_T_224.npz
https://storage.googleapis.com/bit_models/distill/R152x2_T_384.npz
https://github.com/YuanGongND/ast
https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

Posted by: briannguyen @ Aug. 14, 2021, 3:27 p.m.

Pretrained warmup+extra data: https://drive.google.com/file/d/12Kv9S7E3DuRjoFIIPPmlmfELjEV0r-q8/view?usp=sharing

Posted by: KhanhVu @ Aug. 14, 2021, 4:16 p.m.

Phoenix Team bổ sung:
8. GPV (https://github.com/RicherMans/GPV)
Pre: https://github.com/RicherMans/GPV/tree/master/pretrained

9. pyannote (https://huggingface.co/pyannote/segmentation)
Pre: Các model liệt kê tại https://github.com/pyannote/pyannote-audio-hub

10. voxseg (https://github.com/NickWilkinson37/voxseg)
Pre: https://github.com/NickWilkinson37/voxseg/blob/master/voxseg/models/cnn_bilstm.h5

Posted by: yoongaehwa @ Aug. 14, 2021, 4:33 p.m.

Team pretrained model:

- [https://www.tensorflow.org/api_docs/python/tf/keras/applications/efficientnet](https://www.tensorflow.org/api_docs/python/tf/keras/applications/efficientnet)
- [https://tfhub.dev/google/yamnet/1](https://tfhub.dev/google/yamnet/1)
- [https://tfhub.dev/vasudevgupta7/wav2vec2/1](https://tfhub.dev/vasudevgupta7/wav2vec2/1)
- [https://tfhub.dev/google/collections/efficientnet_v2/1](https://tfhub.dev/google/collections/efficientnet_v2/1)
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b0_ra-3dd342df.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b1-533bc792.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b2_ra-bcdf34b7.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b3_ra2-cf984f9c.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b4_ra2_320-7eb33cd5.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth
- https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pth
- [https://github.com/huseinzol05/malaya-speech/tree/master/pretrained-model](https://github.com/huseinzol05/malaya-speech/tree/master/pretrained-model)
- [https://github.com/YuanGongND/ast?fbclid=IwAR2PFNBvZcEFJMmxjgCaeZJGuDAYi7POolqxHrVuHVG8mGQdOpAmVlrGtrY#Pretrained-Models](https://github.com/YuanGongND/ast?fbclid=IwAR2PFNBvZcEFJMmxjgCaeZJGuDAYi7POolqxHrVuHVG8mGQdOpAmVlrGtrY#Pretrained-Models)
- [https://zenodo.org/record/3987831?fbclid=IwAR1IcpESrmeTF65D0kkU43TbrpQvmG2u74nexBuEK3Bo-RQtNWk-tn5p464#.YRfzzvLitaY](https://zenodo.org/record/3987831?fbclid=IwAR1IcpESrmeTF65D0kkU43TbrpQvmG2u74nexBuEK3Bo-RQtNWk-tn5p464#.YRfzzvLitaY)
- "[https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/densenet121_ra-50efcf5c.pth)"
- [https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/2](https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/2)

Posted by: ducquoc.pfs @ Aug. 14, 2021, 4:56 p.m.

Bổ sung
GPV (https://github.com/RicherMans/GPV)
Pre: https://github.com/RicherMans/GPV/tree/master/pretrained
pyannote (https://huggingface.co/pyannote/segmentation)
Pre: Các model liệt kê tại https://github.com/pyannote/pyannote-audio-hub
voxseg (https://github.com/NickWilkinson37/voxseg)
Pre: https://github.com/NickWilkinson37/voxseg/blob/master/voxseg/models/cnn_bilstm.h5

Posted by: ZALO_NGUYEN_QUAN_ANH_MINH @ Aug. 14, 2021, 4:57 p.m.

Xin Chào,

Mình sử dụng pretrained models từ speechbrain (https://speechbrain.readthedocs.io/en/latest/API/speechbrain.pretrained.interfaces.html#summary)

Bao gồm:

* EncoderClassifier: speechbrain/spkrec-ecapa-voxceleb; speechbrain/spkrec-xvect-voxceleb; speechbrain/lang-id-commonlanguage_ecapa
* EncoderDecoderASR: speechbrain/asr-crdnn-transformerlm-librispeech; speechbrain/asr-wav2vec2-commonvoice-en; speechbrain/asr-wav2vec2-transformer-aishell; speechbrain/asr-transformer-aishell; speechbrain/asr-transformer-transformerlm-librispeech
* Sepformer: speechbrain/sepformer-whamr
* SpectralMaskEnhancement: speechbrain/metricgan-plus-voicebank; speechbrain/mtl-mimic-voicebank

Bổ sung thêm 2 models từ tf.keras.application:

* tf.keras.applications.efficientnet.EfficientNetB7
* tf.keras.applications.vgg19.VGG19

Posted by: trungnt13 @ Aug. 14, 2021, 5:13 p.m.

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pth
https://github.com/harritaylor/torchvggish

Posted by: longgage @ Aug. 14, 2021, 8:11 p.m.

https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b0_aa-827b6e33.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b1_aa-ea7a6ee0.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b2_aa-60c94f97.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_efficientnet_b3_aa-84b4657e.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pth
https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mixnet_s-a907afbc.pth
https://github.com/harritaylor/torchvggish

Posted by: oggyfaker @ Aug. 14, 2021, 8:15 p.m.


https://github.com/YuanGongND/ast
https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

Posted by: tinyswish @ Aug. 15, 2021, 1:03 a.m.

Bổ sung
https://github.com/YuanGongND/ast
https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1
Và các pretrained model của timm 0.4.12

Posted by: meoconxinhxan @ Aug. 23, 2021, 9:19 a.m.

Additionally,
PANNS: https://zenodo.org/record/3987831#.YRfbe4gplPa

Posted by: nhattruong @ Aug. 23, 2021, 9:39 a.m.

Bổ Sung
https://zenodo.org/record/3987831#.YSO3QnUzaV4

Posted by: longgage @ Aug. 23, 2021, 2:57 p.m.

Bổ Sung
https://github.com/YuanGongND/ast
https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1
https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1
https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1
https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1
https://zenodo.org/record/3987831#.YSO3qXUzaV5
https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1
https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1
https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1

Posted by: oggyfaker @ Aug. 23, 2021, 2:59 p.m.
Post in this thread