MycroftAI
/
mycroft-precise
mirror of https://github.com/MycroftAI/mycroft-precise.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
268 Commits
8 Branches
4 Tags
735 KiB
Commit Graph

268 Commits (cfc36aafb22536de714507cc5738ad07aa494b77)

Author SHA1 Message Date
Matthew Scholefield 02fa053653 Fix loss bias in trainer.py 2018-08-10 00:02:39 -05:00
Matthew Scholefield 63b4592675 Fix activations_per_day in precise-simulate 2018-08-10 00:01:16 -05:00
Matthew Scholefield e43498d2ea Explicitly delete large vectors 
This reduces potential memory overlap
 2018-08-10 00:00:30 -05:00
Matthew Scholefield b1d7a7ff5d Show percentage when calculating vectors 2018-08-09 23:59:26 -05:00
Matthew Scholefield 7bded13de4 Change vectorizer to Sonopy 
Also adds vectorizer parameter to choose between mels and mfccs
 2018-08-09 23:58:21 -05:00
Matthew Scholefield 754d332b10 Reorganize and improve setup.py 
Added keywords and license keys
 2018-08-09 23:10:59 -05:00
Matthew Scholefield d71ea3fee0 Add visual element to precise-listen 
The previous default `precise-listen` is not `precise-listen -l`
 2018-08-09 11:36:10 -05:00
Matthew D. Scholefield 8cfbd41d8c
Fix Python 3.4 compatibility in train-incremental 2018-08-07 16:27:56 -05:00
Matthew D. Scholefield fd2b71b069
Replace Python 3.5+ syntax 
Otherwise it errors on Python 3.4
 2018-08-06 16:54:26 -05:00
Matthew Scholefield 7b802be7b8 Move Trainer into precise.scripts.train 
Not having it there was inconsistent because the other trainer classes are in the scripts package within the respective scripts
 2018-07-26 13:19:10 -05:00
Matthew Scholefield 9642fcc07f Inject params in precise-simulate 2018-07-25 10:06:14 -05:00
Matthew D. Scholefield d17002cd4f
Update training link 2018-07-17 14:47:46 -05:00
Mikhail Dmitrienko 81a8b18986 Removed redundant code (#20) 
Removed a redundant line of code. print("False positive: ", false_positives * 100, "%") is already printed so print(false_positives) isn't needed.
 2018-07-16 15:22:00 -05:00
Mikhail Dmitrienko 75e0769026 Changed the maximum possible amount of units (#19) 
Changed the maximum possible amount of units for the optimization to return to 70 units.
 2018-07-16 11:57:43 -05:00
Matthew Scholefield b7ae1472f1 Upgrade requirements.txt 2018-07-13 14:52:12 -05:00
Matthew Scholefield d6db6c8ec2 Make train_optimize inherit from Trainer 2018-07-12 21:22:08 -05:00
Matthew D. Scholefield 8145dedefa
Merge pull request #18 from mikhaildmitrienko/hyperparameter 
Created train_optimizer.py
 2018-07-12 17:27:01 -05:00
Matthew Scholefield 651901143d Add precise-train-optimize alias for train_optimize.py 2018-07-12 17:24:01 -05:00
Matthew Scholefield 7f915f939d Remove unused variables 2018-07-12 17:21:54 -05:00
Matthew Scholefield 3549765a84 Move module level code into main function 2018-07-12 17:21:54 -05:00
Matthew Scholefield f917a72f4c Add bbopt to dependencies 2018-07-12 17:21:54 -05:00
Matthew Scholefield 4d1f212c2b Remove environment specific definition 2018-07-12 17:21:54 -05:00
Matthew Scholefield c5dd7c891d Autoformat and remove unused imports 2018-07-12 17:21:54 -05:00
Matthew Scholefield 5818a101da Add license header 2018-07-12 17:21:54 -05:00
Matthew Scholefield 4f5df1b997 Make imports absolute 2018-07-12 17:21:54 -05:00
Matthew Scholefield ac43e4b6bb Add .py extension to train_optimize 
Also rename to train_optimize to be consistent with verbiage
 2018-07-12 17:21:54 -05:00
mikhaildmitrienko 01dc86896d Created train_optimizer.py 2018-07-12 17:21:54 -05:00
Matthew Scholefield 7d97cd2331 Add simulate script 2018-07-12 16:57:02 -05:00
Matthew Scholefield 04174efeb8 Remove executable flag from setup.py 2018-07-12 14:43:05 -05:00
Matthew Scholefield d32fb6fee3 Fix extra comma on assignment 2018-07-12 14:21:54 -05:00
Matthew Scholefield 8696ef80ba Extract TriggerDetector class to simplify PreciseRunner 
This also makes it easier to perform similar functionality in an upcoming simulationn script
 2018-07-12 13:38:39 -05:00
Matthew Scholefield 36804075b3 Fix Nan in false_neg and false_pos metrics 
This was caused when small batches with no positive samples caused a divide-by-zero
 2018-07-11 19:34:22 -05:00
Matthew Scholefield d8dd0f3240 Shuffle training and testing data before training 
This fixes issues with small batch sizes and metrics
 2018-07-11 19:32:39 -05:00
Matthew Scholefield 06d9669e79 Add sampled training support 2018-07-11 12:45:41 -05:00
Matthew Scholefield 1f5e17ba0a Formatting 2018-07-11 12:45:41 -05:00
Matthew Scholefield afbace91b1 Add logs to .gitignore 2018-07-11 12:45:41 -05:00
Matthew Scholefield 6a8b7b8ea8 Add delta vector support 2018-07-11 12:45:41 -05:00
Matthew Scholefield ebd5e09feb Rework train script to use new Trainer class 
This allows IncrementalTrainer to share similar behavior
 2018-07-11 12:45:41 -05:00
Matthew Scholefield 4da65d1f37 Quiet tensorflow when performing predictions 2018-07-11 12:45:41 -05:00
Matthew Scholefield a946de21ce Make ListenerParams proper class via attr 2018-07-11 12:45:41 -05:00
Matthew Scholefield 779ddd2d02 Use vectorize_raw instead of rewriting mfcc call 2018-07-11 12:45:41 -05:00
Matthew Scholefield 0da11b33bd Make mfcc generation use mfcc_size instead of feature_size 
feature_size could be different.
 2018-07-11 12:45:41 -05:00
Matthew Scholefield 603385291f Fix up docstring for create_model 2018-07-11 12:45:41 -05:00
Matthew Scholefield 252abbeb15 Make loss bias use module level variable 2018-07-11 12:45:41 -05:00
Matthew D. Scholefield 3fc1fe4789 Add libatlas-base-dev to debian dependencies 
It is a runtime dependency of the piwheels tensorflow whl
 2018-07-02 20:41:54 +00:00
Matthew Scholefield 1f15f9da92 Increase default batch size in train-incremental 2018-06-25 12:37:56 -05:00
Matthew D. Scholefield a83769302f Remove dist/ folder from tar 2018-06-22 10:38:34 -05:00
Matthew Scholefield c663719cce Increment version to 0.2.0 2018-06-20 16:49:50 -05:00
Wu Zhangjin 523eb719f1 listen: not-wake-up: increase the random value's range 
the old range is too narrow and therefore make data override often happen.

Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
 2018-06-20 16:49:50 -05:00
Wu Zhangjin 83bd6804c9 README: replace precise-record with precise-listen 
sync with commit 19bf49a4a.

Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
 2018-06-20 16:49:50 -05:00