Development

CONTRIBUTING.md

@@ -118,17 +118,18 @@ version: 0.0.1         # Function version, should follow the standard semantic v
 The table below represents the available options to choose from when filling `item.yaml`'s `categories` property. Please 
 select one or more categories and feel free to suggest additional options.
 
-| Name | How should it be written in item.yaml | When should i choose this category? |
-|------|---------------------------------------|-------------------------------------|
-| Data Preparation | <b>data-preparation</b> | A function that prepares, cleans or otherwise processes data in any other number of ways |
-| Extract, Transform, Load | <b>etl</b> | A function that can be used to take data from some source, transforms it and finally load it to some destination |
-| Machine Learning | <b>machine-learning</b> | A function that is related to machine learning in any way, can be used along-side categories like Data Preparation, Model Training, Model Serving |
-| Model Serving | <b>model-serving</b> | A function that can be used to serve models via V2ModelServer |
-| Model Training | <b>model-training</b> | A function that can be used for training Machine Learning models |
-| Model Testing | <b>model-testing</b> | A function that can be used to test trained models |
-| Data Analysis | <b>data-analysis</b> | A function that can be used to preform analysis and exploration on data |
-| Monitoring | <b>monitoring</b> | A function that can be used to monitor and/or measure metrics regarding data, models and model endpoints |
-| Utilities | <b>Utilities</b> | A function that is used to alert, notify or provide services in any way |
+| Name                     | How should it be written in item.yaml | When should i choose this category? |
+|--------------------------|---------------------------------------|-------------------------------------|
+| Data Preparation         | <b>data-preparation</b>               | A function that prepares, cleans or otherwise processes data in any other number of ways |
+| Extract, Transform, Load | <b>etl</b>                            | A function that can be used to take data from some source, transforms it and finally load it to some destination |
+| Machine Learning         | <b>machine-learning</b>               | A function that is related to machine learning in any way, can be used along-side categories like Data Preparation, Model Training, Model Serving |
+| Model Serving            | <b>model-serving</b>                  | A function that can be used to serve models via V2ModelServer |
+| Model Training           | <b>model-training</b>                 | A function that can be used for training Machine Learning models |
+| Model Testing            | <b>model-testing</b>                  | A function that can be used to test trained models |
+| Data Analysis            | <b>data-analysis</b>                  | A function that can be used to preform analysis and exploration on data |
+| Deep Learning            | <b>deep-learning</b>                  | A function that can be used to teaches computers to process data in a way that is inspired by the human brain |
+| Monitoring               | <b>monitoring</b>                     | A function that can be used to monitor and/or measure metrics regarding data, models and model endpoints |
+| Utilities                | <b>Utilities</b>                      | A function that is used to alert, notify or provide services in any way |
 
 ## Installation Guide
 It is highly advised using the functions package in a dedicated environment, since Pipenv is used as part of the testing routine, 

pyannote_audio/function.yaml

@@ -2,12 +2,12 @@ kind: job
 metadata:
   name: pyannote-audio
   tag: ''
-  hash: 335752327ddd14b62222bd45faa3a88704505b66
+  hash: c45be8d7f51f0b2203155b08c307814a2cb0ac78
   project: ''
   labels:
     author: guyl
   categories:
-  - Deep Learning
+  - deep-learning
   - Huggingface
   - Audio
 spec:
@@ -35,24 +35,27 @@ spec:
       - name: root_worker_inputs
         type: Dict[str, Any]
         default: null
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 61
+      has_varargs: false
+      has_kwargs: false
     decorator:
       name: decorator
       doc: ''
       parameters:
       - name: handler
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 73
+      has_varargs: false
+      has_kwargs: false
     wrapper:
       name: wrapper
       doc: ''
       parameters: []
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 78
+      has_varargs: false
+      has_kwargs: true
     diarize:
       name: diarize
       doc: "Perform speech diarization on given audio files using pyannote-audio (https://github.com/pyannote/pyannote-audio).\n\
@@ -131,8 +134,10 @@ spec:
         default: false
       outputs:
       - doc: 'A tuple of:'
-        default: ''
+        type: Tuple[Dict[str, List[Tuple[float, float, str]]], Dict[str, str]]
       lineno: 139
+      has_varargs: false
+      has_kwargs: false
   description: pyannote's speech diarization of audio files
   default_handler: diarize
   disable_auto_mount: false

pyannote_audio/item.yaml

@@ -1,8 +1,8 @@
 apiVersion: v1
 categories:
-  - Deep Learning
-  - Huggingface
-  - Audio
+- deep-learning
+- Huggingface
+- Audio
 description: pyannote's speech diarization of audio files
 doc: ''
 example: pyannote_audio.ipynb
@@ -22,9 +22,9 @@ spec:
   image: mlrun/mlrun-gpu
   kind: job
   requirements:
-    - pyannote.audio
-    - pyannote.core
-    - torchaudio
-    - tqdm
+  - pyannote.audio
+  - pyannote.core
+  - torchaudio
+  - tqdm
 url: ''
-version: 1.0.0
+version: 1.1.0

silero_vad/function.yaml

@@ -2,12 +2,12 @@ kind: job
 metadata:
   name: silero-vad
   tag: ''
-  hash: bc0ad5572cc391fcdc93baaee48e1ef949a7984d
+  hash: 61b7a70c167b7819481fdabf9350fc6fa344d2f5
   project: ''
   labels:
     author: guyl
   categories:
-  - Deep Learning
+  - deep-learning
   - PyTorch
   - Audio
 spec:
@@ -34,8 +34,9 @@ spec:
       outputs:
       - doc: The audio file of the task.
         type: Path
-        default: ''
       lineno: 43
+      has_varargs: false
+      has_kwargs: false
     do_task:
       name: do_task
       doc: Do the task on the given speech timestamps. The task will diarize the VAD
@@ -46,18 +47,21 @@ spec:
         type: List[List[Dict[str, int]]]
         doc: The speech timestamps per channel to do the task on as outputted from
           the VAD.
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 94
+      has_varargs: false
+      has_kwargs: false
     get_result:
       name: get_result
       doc: Get the result of the task. A tuple of the audio file name and the result.
       parameters:
       - name: self
       outputs:
       - doc: The result of the task.
-        default: ''
+        type: Tuple[str, list]
       lineno: 61
+      has_varargs: false
+      has_kwargs: false
     to_tuple:
       name: to_tuple
       doc: Convert the task to a tuple to reconstruct it later (used for multiprocessing
@@ -66,8 +70,10 @@ spec:
       - name: self
       outputs:
       - doc: The converted task.
-        default: ''
+        type: Tuple[str, dict]
       lineno: 116
+      has_varargs: false
+      has_kwargs: false
     create_task:
       name: create_task
       doc: Create a task with the given audio file.
@@ -79,8 +85,9 @@ spec:
       outputs:
       - doc: The created task.
         type: BaseTask
-        default: ''
       lineno: 146
+      has_varargs: false
+      has_kwargs: false
     from_tuple:
       name: from_tuple
       doc: Create a task from a tuple of the audio file name and the task kwargs.
@@ -92,8 +99,9 @@ spec:
       outputs:
       - doc: The created task.
         type: BaseTask
-        default: ''
       lineno: 157
+      has_varargs: false
+      has_kwargs: false
     load:
       name: load
       doc: Load the VAD model.
@@ -104,9 +112,10 @@ spec:
         doc: Whether to force reload the model even if it was already loaded. Default
           is True.
         default: true
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 234
+      has_varargs: false
+      has_kwargs: false
     detect_voice:
       name: detect_voice
       doc: "Perform voice activity detection on given audio files using the silero\
@@ -186,9 +195,10 @@ spec:
         type: bool
         doc: Verbosity.
         default: false
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 393
+      has_varargs: false
+      has_kwargs: false
     diarize:
       name: diarize
       doc: "Perform speech diarization on given audio files using the silero VAD model\
@@ -264,9 +274,10 @@ spec:
         type: bool
         doc: Verbosity.
         default: false
-      outputs:
-      - default: ''
+      outputs: []
       lineno: 517
+      has_varargs: false
+      has_kwargs: false
   description: Silero VAD (Voice Activity Detection) functions.
   default_handler: detect_voice
   disable_auto_mount: false

silero_vad/item.yaml

@@ -1,8 +1,8 @@
 apiVersion: v1
 categories:
-  - Deep Learning
-  - PyTorch
-  - Audio
+- deep-learning
+- PyTorch
+- Audio
 description: Silero VAD (Voice Activity Detection) functions.
 doc: ''
 example: silero_vad.ipynb
@@ -22,9 +22,9 @@ spec:
   image: mlrun/mlrun
   kind: job
   requirements:
-    - torch
-    - torchaudio
-    - tqdm
-    - onnxruntime
+  - torch
+  - torchaudio
+  - tqdm
+  - onnxruntime
 url: ''
-version: 1.1.0
+version: 1.2.0