search

qiuqiangkong / DCASE2016_Task4

4 stars 3 forks source link

Eval dataset #1

Closed kelelexu closed 5 years ago

kelelexu commented 6 years ago

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

qiuqiangkong commented 6 years ago

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task-large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

― You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl.

kelelexu commented 6 years ago

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong notifications@github.com wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

― You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

qiuqiangkong commented 6 years ago

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task-large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong notifications@github.com wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

kelelexu commented 6 years ago

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations ' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong notifications@github.com wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69eZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

qiuqiangkong commented 6 years ago

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations ' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong notifications@github.com wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69eZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

kelelexu commented 6 years ago

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/ chunk_annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong notifications@github.com

wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/ issues/1#issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv- 69eZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

kelelexu commented 6 years ago

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong notifications@github.com

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

qiuqiangkong commented 6 years ago

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong notifications@github.com

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

kelelexu commented 6 years ago

There are wav files of 48 KHZ, wav files of 1KHZ, and the csv files. There are 18411 files in total, 3 * 6137.

On Sat, Nov 25, 2017 at 9:00 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong notifications@github.com

wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/ dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939221, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZvyl-UeEre5XsNMfjBY9XKoU59ctmks5s6A-BgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

qiuqiangkong commented 6 years ago

Hi Kele,

You may look into development_chunks_raw.csv and evaluation_chunks_raw.csv. I remember around 4000 are for development and around 1900 for evaluation.

Please let me know if there are any questions.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 13:07:09 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

There are wav files of 48 KHZ, wav files of 1KHZ, and the csv files. There are 18411 files in total, 3 * 6137.

On Sat, Nov 25, 2017 at 9:00 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong notifications@github.com

wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/ dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939221, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZvyl-UeEre5XsNMfjBY9XKoU59ctmks5s6A-BgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939549, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5yXPvbDEQ0sEl4T5buMmzYjLRahaKks5s6BD9gaJpZM4QoxLl.

kelelexu commented 6 years ago

OK, thanks so much!~

On Sat, Nov 25, 2017 at 9:12 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

You may look into development_chunks_raw.csv and evaluation_chunks_raw.csv. I remember around 4000 are for development and around 1900 for evaluation.

Please let me know if there are any questions.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 13:07:09 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

There are wav files of 48 KHZ, wav files of 1KHZ, and the csv files. There are 18411 files in total, 3 * 6137.

On Sat, Nov 25, 2017 at 9:00 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/ dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_ jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub< https://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1>, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHub< https://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395>, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/ issues/1#issuecomment- 346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346939221, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZvyl- UeEre5XsNMfjBY9XKoU59ctmks5s6A-BgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939549, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5yXPvbDEQ0sEl4T5buMmzYjLRahaKks5s6BD9gaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939808, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv9AYFSGRU6qqfg2kDIEca95Zo-VDks5s6BI4gaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

kelelexu commented 6 years ago

大佬能不能加个QQ

On Sat, Nov 25, 2017 at 9:12 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

You may look into development_chunks_raw.csv and evaluation_chunks_raw.csv. I remember around 4000 are for development and around 1900 for evaluation.

Please let me know if there are any questions.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 13:07:09 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

There are wav files of 48 KHZ, wav files of 1KHZ, and the csv files. There are 18411 files in total, 3 * 6137.

On Sat, Nov 25, 2017 at 9:00 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/ dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_ jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub< https://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1>, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHub< https://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395>, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/ issues/1#issuecomment- 346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346939221, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZvyl- UeEre5XsNMfjBY9XKoU59ctmks5s6A-BgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939549, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5yXPvbDEQ0sEl4T5buMmzYjLRahaKks5s6BD9gaJpZM4QoxLl.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939808, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv9AYFSGRU6qqfg2kDIEca95Zo-VDks5s6BI4gaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

qiuqiangkong commented 6 years ago

Hi Kele,

Please feel free to add my WeChat "richard2357".

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 20 January 2018 07:28:09 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

大佬能不能加个QQ

On Sat, Nov 25, 2017 at 9:12 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

You may look into development_chunks_raw.csv and evaluation_chunks_raw.csv. I remember around 4000 are for development and around 1900 for evaluation.

Please let me know if there are any questions.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 13:07:09 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

There are wav files of 48 KHZ, wav files of 1KHZ, and the csv files. There are 18411 files in total, 3 * 6137.

On Sat, Nov 25, 2017 at 9:00 PM, qiuqiangkong notifications@github.com wrote:

Hi Kele,

What and how many files are there in chunks/?

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 11:24:10 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi, Qiuqiang, Sorry to disturb you again. I downloaded the dataset. And all of the files are given as follows:

CHANGES chunks/ development_chunks_raw.csv development_chunks_refined.csv development_chunks_refined_crossval_dcase2016.csv evaluation_chunks_raw.csv evaluation_chunks_refined.csv LICENSE README VERSION

Here, I can't find the eva_chunks..

Hope to get your help.

Thanks so much!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:50 PM, Kele XU kelele.xu@gmail.com wrote:

Hi, Qiuqiang,

Thanks!~

Best Regards, Kele

On Sat, Nov 25, 2017 at 6:26 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

The CHiME home dataset including development and evaluation can be downloaded here http://www.cs.tut.fi/sgn/arg/ dcase2016/task-audio-tagging

Please let me know if there are any questions!

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 25 November 2017 10:13:01 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi Qiuqiang,

Thanks so much for your feedback. In fact, I am just starting to use DCASE dataset. I found your code which is quite helpful for me. In the link https://github.com/qiuqiangkong/audio_tagging_ jdc/blob/master/config.py

when I am trying to config my local environment, I cannot find the eva_chunks in the Chime_home dataset. Could you please give me some suggestions? Thanks so much.

Kele Best Wishes!

evaluation set

eva_wav_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/eva_chunks' eva_anno_fd = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk_annotations/annotations

' eva_csv_path = ' /vol/vssp/AP_datasets/audio/dcase2016/task4/chime_home/chunk _annotations/evaluation_chunks_refined.csv '

On Thu, Nov 23, 2017 at 11:19 PM, qiuqiangkong < notifications@github.com>

wrote:

Hi Kele,

Both the development & evaluation (863 Mb) set can be downloaded from http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection. However, the labels are not released by the organizers yet. Please let us know if there is any problems in downloading the data.

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 14:42:51 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong Q Mr (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Thanks so much for your promptly reply. In fact, I wonder where can we get the eval dataset for DCASE 2016 TASK 4?

On Thu, Nov 23, 2017 at 10:31 PM, qiuqiangkong < notifications@github.com> wrote:

Hi,

Many thanks for the interest!

The evaluation dataset can be found on the official website of DCASE 2017 Task 4: http://www.cs.tut.fi/sgn/arg/dcase2017/challenge/task- large-scale-sound-event-detection

Best wishes,

Qiuqiang

From: KeleXU notifications@github.com Sent: 23 November 2017 13:41:26 To: qiuqiangkong/DCASE2016_Task4 Cc: Subscribed Subject: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

I wonder where can be get the eval dataset? Thanks so much for your feedback.

谢谢大佬

D You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub< https://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1>, or mute the thread< https://github.com/notifications/unsubscribe-auth/ AMt5ydclVUYhbCxZQP9IKDHId_9ekAibks5s5XYGgaJpZM4QoxLl>.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment- 346633059, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv3y05QNynFr30OOfgQkZGSEoONM_ks5s5YHagaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346635619, or mute the threadhttps://github.com/notifications/unsubscribe- auth/AMt5yYKY3X3ph4I0DlfaTU-cqdKbj1O2ks5s5YRqgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1# issuecomment-346644564, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv-69e ZDNfCHCEmBiPUzRAL2oldPKks5s5Y0KgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHub< https://github.com/qiuq iangkong/DCASE2016_Task4/issues/1#issuecomment-346931395>, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt 5ySFaT68IRyVLsDg5UrBwDbh5VQJGks5s5-gsgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/ issues/1#issuecomment- 346932043, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHaZv1ODce6paQ1MMAsvGzHNmJcTXmq_ks5s5-tzgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346934652, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5ySnsLyraAihxZHaeDgnCW2JzK35cks5s5_jZgaJpZM4QoxLl.

D You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment- 346939221, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZvyl- UeEre5XsNMfjBY9XKoU59ctmks5s6A-BgaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

D You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/ qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939549, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AMt5yXPvbDEQ0sEl4T5buMmzYjLRahaKks5s6BD9gaJpZM4QoxLl.

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-346939808, or mute the thread https://github.com/notifications/unsubscribe-auth/AHaZv9AYFSGRU6qqfg2kDIEca95Zo-VDks5s6BI4gaJpZM4QoxLl .

-- Kele Xu, Institute Langevin ESPCI 75005 PARIS FRANCE

― You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-359152507, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5ySJWwMBioL7NguGVRw1vuQTvuHzgks5tMZWJgaJpZM4QoxLl.

shaurya2904 commented 5 years ago

Hi qiugiangkong, I found your JDC code quite helpful although setting up the required environment for it was quite tricky on conda. After running prepare_dev_data.py. While running the code main code it gives the non type error for main_eva_jdc and attribute error for 8th layer a8.shape [ No attribute shape exist] for main_evo_bob.py.

I gather both the above errors are due to lack of data at input layer which hence is not passing it to subsequent layers. . Which I think should be generated by prepare_dev_data.py.

Can you please help me out?

qiuqiangkong commented 5 years ago

Hi, we are using DCASE 2016 Task 4 dataset: http://www.cs.tut.fi/sgn/arg/dcase2016/task-audio-tagging

Please let us know if there are further questions.

Best wishes,

Qiuqiang

From: shaurya2904 notifications@github.com Sent: 08 January 2019 03:17:17 To: qiuqiangkong/DCASE2016_Task4 Cc: Kong, Qiuqiang (PG/R - Elec Electronic Eng); Comment Subject: Re: [qiuqiangkong/DCASE2016_Task4] Eval dataset (#1)

Hi qiugiangkong, I found your JDC code quite helpful although setting up the required environment for it was quite tricky on conda. After running prepare_dev_data.py. While running the code main code it gives the non type error for main_eva_jdc and attribute error for 8th layer a8.shape [ No attribute shape exist].

I gather both the above errors are due to lack of an image( spectrogram) . Which I think should be generated by prepare_dev_data.py but due to lack of eva_chunk folder and annotation file in chime dataset I downloaded this is occuring.

Can you please link me to proper dataset?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/qiuqiangkong/DCASE2016_Task4/issues/1#issuecomment-452161220, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMt5yWwpGll8LB_sBmi6g6gJqlmFmHCyks5vBA29gaJpZM4QoxLl.

shaurya2904 commented 5 years ago

Thank you fro prompt reply I am using the dataset you have mentioned above and using the unaltered codes in audio_tagging_jdc repository. But it is still leading to NoneType error. Which I am unable to pin point as to what in the original code is preventing the prepared data from getting pushed to the layer.

The error is occuring while calling getloss function. For instance in main_dev_bob.py line 48 (n_songs, n_chunk, n_out) = a8_node.shape is generating

AttributeError: 'NoneType' object has no attribute 'shape'

I wanted to share screenshot of the error. If there can be a easier correspondence with you it would be really helpful.

Regards, Shaurya2904