Commits · 39d55d9a4b3e5b2d894e51e78d9fa35e6be4aa18 · ModelZoo / ResNet50_tensorflow

13 Jan, 2020 3 commits
- Tolerate missing soundfile module. (#8039) · 39d55d9a
  Dan Ellis authored Jan 13, 2020
```
* Update slim include.

* Force installation of tensorflow 1.14 (not tf2).

* Code will run without soundfile (but die if you attempt to read a sound file).
```
  39d55d9a
- Update slim include. (#8036) · e9233425
  Dan Ellis authored Jan 13, 2020
```
* Update slim include.

* Force installation of tensorflow 1.14 (not tf2).
```
  e9233425
- Update VGGish README for new path, soundfile (#8038) · 8b479cc7
  Dan Ellis authored Jan 13, 2020
  
  8b479cc7
21 Nov, 2019 1 commit
- Adding research/audioset/yamnet, a pre-trained audio event classifier (#7850) · dfffd623
  Dan Ellis authored Nov 21, 2019
```
Add files for YamNet, a sound event classifier.
```
  dfffd623
13 Jun, 2019 1 commit
- Moved research/audioset VGGish code into its own subdirectory (#7009) · 4079c5d9
  Manoj Plakal authored Jun 13, 2019
```
* Moved VGGish code into its own directory.

* Moved most of old README.md into vggish/README.md.
```
  4079c5d9
12 Nov, 2018 1 commit
- Shortened information about used library versions in Audioset README · de51e746
  Samuel Neugber authored Nov 12, 2018
  
  de51e746
08 Nov, 2018 1 commit
- Switching to more robust pysoundfile for reading wav files · dec81ac7
  Samuel Neugber authored Nov 07, 2018
  
  dec81ac7
05 Nov, 2018 1 commit
- Revert "Use explicit relative import syntax for python 3 compatibility in audioset." · 7d2da5cb
  David T.H. Kao authored Nov 05, 2018
  
  7d2da5cb
04 Nov, 2018 1 commit
- Use explicit relative import syntax for python 3 compatibility. · f54d201d
  David Kao authored Nov 03, 2018
  
  f54d201d
20 Sep, 2018 1 commit
- Fixed typo in --train_vggish flag description. · 58877553
  Souradip Mookerjee authored Sep 02, 2018
  
  58877553
15 Aug, 2018 1 commit

Add a pointer to a Colab for the embeddings · 60d6808b

Malcolm Slaney authored Aug 14, 2018

Added a pointer to a Colab that illustrates how to use the public AudioSet models to generate embeddings for user-specified sounds.

60d6808b

13 Feb, 2018 3 commits
- Cosmetic changes to spectrogram_to_mel_matrix() · bdf0f628
  Manoj Plakal authored Feb 13, 2018
```
To simplify syncing this c ode back into Google.
```
  bdf0f628
- Fix double paste. · 982a5504
  Manoj Plakal authored Feb 13, 2018
  
  982a5504
- Add sanity checks to mel computation. · 4b411030
  Manoj Plakal authored Feb 13, 2018
```
Clamp the requested frequency range to [0, Nyquist].
```
  4b411030
09 Feb, 2018 1 commit

Explicitly specify checkpoint version. · d9f6b6f3

Manoj Plakal authored Feb 09, 2018


TF Saver now requires specifying the checkpoint version even when restoring to avoid errors when we specify
a path to a checkpoint file instead of a directory.

d9f6b6f3

15 Oct, 2017 1 commit
- Fixing wrong path · d72df8b0
  Quentin Pleplé authored Oct 15, 2017
  
  d72df8b0
12 Oct, 2017 1 commit
- Concatenate instead of add in vggish_train_demo.py. · a7c84b82
  Manoj Plakal authored Oct 12, 2017
```
Also added 'sudo' to installation instructions.
```
  a7c84b82
23 Sep, 2017 1 commit
- Fix broken links in the models repo (#2445) · 4a705e08
  Neal Wu authored Sep 22, 2017
  
  4a705e08
21 Sep, 2017 1 commit
- Move the research models into a research subfolder (#2430) · f87a58cd
  Neal Wu authored Sep 21, 2017
  
  f87a58cd