You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<i>mlflow.js</i> is an open-source npm library designed for JavaScript developers who want to integrate with MLflow, providing tools and functionalities for managing machine learning lifecycle.
15
-
16
-
<ahref="https://www.mlflow-js.org/">Visit the official <i>mlflow.js</i> site for more info!</a>
<i>MLflow.js</i> is an open-source JavaScript library that helps developers track machine learning experiments and manage models with MLflow, providing functionalities for machine learning lifecycle in JavaScript/TypeScript environments.
21
15
22
16
<br>
23
17
24
18
## Features
25
19
26
-
<i>mlflow.js</i> covers all REST API endpoints under MLflow's Tracking Server and Model Registry. Official documentation for <i>mlflow.js</i> can be found <ahref="https://www.mlflow-js.org/documentation">here</a>. Moreover, high-level abstraction workflows have been developed to facilitate developers' work processes.
27
-
28
-
### High-Level Abstraction Workflows
29
-
30
-
**Experiment Manager**
31
-
32
-
- runExistingExperiment - Full workflow of creating, naming, and starting a run under an existing experiment, logging metrics, params, tags, and the model, and finishing the run
33
-
- runNewExperiment - Full workflow of creating, naming, and starting a run under a new experiment, logging metrics, params, tags, and the model, and finishing the run
34
-
- experimentSummary - Returns an array of all the passed-in experiment's runs, sorted according to the passed-in metric
20
+
<i>MLflow.js</i> covers all REST API endpoints under MLflow's Tracking Server and Model Registry. Moreover, high-level abstractions have been developed to facilitate developers' common ML workflows. It provides some key advantages:
35
21
36
-
**Run Manager**
37
-
38
-
- cleanupRuns - Deletes runs that do not meet certain criteria and return an object of deleted runs and details
39
-
- copyRun - Copies a run from one experiment to another (without artifacts and models)
40
-
41
-
**Model Manager**
42
-
43
-
- createRegisteredModelWithVersion - Creates a new registered model and the first version of that model
44
-
- updateRegisteredModelDescriptionAndTag - Updates a registered model's description and tags
45
-
- updateAllLatestModelVersion - Updates the latest version of the specified registered model's description, adds a new alias, and tag key/value for the latest version
46
-
- setLatestModelVersionTag - Adds a new tag key/value for the latest version of the specified registered model
47
-
- setLatestModelVersionAlias - Adds an alias for the latest version of the specified registered model
48
-
- updateLatestModelVersion - Updates the description of the latest version of a registered model
49
-
- updateAllModelVersion - Updates the specified version of the specified registered model's description and adds a new alias and tag key/value for that specified version
50
-
- deleteLatestModelVersion - Deletes the latest version of the specified registered model
51
-
- createModelFromRunWithBestMetric - Creates a new model with the specified model name from the run with the best specified metric
- Type Safety: Built with TypeScript for enhanced developer experience and code reliability
24
+
- Modular Architecture: Designed with object-oriented structure that mirrors MLflow's concepts while being extensible and maintainable
25
+
- Client-side ML Compatibility: Complements popular JavaScript libraries like TensorFlow.js, enabling ML deployment directly in the browser or client side
52
26
53
27
<br>
54
28
@@ -71,7 +45,7 @@ Visit our LinkedIn page below:
71
45
72
46
## Prerequisites
73
47
74
-
### Set Up MLflow UI
48
+
### Set Up MLflow
75
49
76
50
Ensure MLflow is installed on your system:
77
51
@@ -91,6 +65,9 @@ mlflow ui --port 5000
91
65
92
66
This will launch the MLflow UI on your local machine at `http://localhost:5000`.
93
67
68
+
### Development Setup
69
+
For development environment setup instructions, please refer to our [Contributing Guide](../CONTRIBUTING.md).
70
+
94
71
<br>
95
72
96
73
## Quickstart
@@ -122,18 +99,49 @@ async function createExperiment(){
-[Quick Tutorials](https://www.mlflow-js.org/) - Video walkthrough of the example code with MLflow UI
107
+
-[Read our Medium Article](link) - Overview on why we built MLflow.js and how it enhances ML workflows in JavaScript environments
125
108
126
109
<br>
127
110
128
111
## Documentation
129
112
130
-
Official documentation for <i>mlflow.js</i> can be found <ahref="https://www.mlflow-js.org/documentation">here</a>.
113
+
Official documentation for <i>MLflow.js</i> can be found <ahref="https://www.mlflow-js.org/documentation">here</a>.
114
+
115
+
### High-Level Workflows
116
+
117
+
**Experiment Manager**
118
+
119
+
- runExistingExperiment - Full workflow of creating, naming, and starting a run under an existing experiment, logging metrics, params, tags, and the model, and finishing the run
120
+
- runNewExperiment - Full workflow of creating, naming, and starting a run under a new experiment, logging metrics, params, tags, and the model, and finishing the run
121
+
- experimentSummary - Returns an array of all the passed-in experiment's runs, sorted according to the passed-in metric
122
+
123
+
**Run Manager**
124
+
125
+
- cleanupRuns - Deletes runs that do not meet certain criteria and return an object of deleted runs and details
126
+
- copyRun - Copies a run from one experiment to another (without artifacts and models)
127
+
128
+
**Model Manager**
129
+
130
+
- createRegisteredModelWithVersion - Creates a new registered model and the first version of that model
131
+
- updateRegisteredModelDescriptionAndTag - Updates a registered model's description and tags
132
+
- updateAllLatestModelVersion - Updates the latest version of the specified registered model's description, adds a new alias, and tag key/value for the latest version
133
+
- setLatestModelVersionTag - Adds a new tag key/value for the latest version of the specified registered model
134
+
- setLatestModelVersionAlias - Adds an alias for the latest version of the specified registered model
135
+
- updateLatestModelVersion - Updates the description of the latest version of a registered model
136
+
- updateAllModelVersion - Updates the specified version of the specified registered model's description and adds a new alias and tag key/value for that specified version
137
+
- deleteLatestModelVersion - Deletes the latest version of the specified registered model
138
+
- createModelFromRunWithBestMetric - Creates a new model with the specified model name from the run with the best specified metric
131
139
132
140
<br>
133
141
134
142
## Contributing
135
143
136
-
We welcome contributions to <i>mlflow.js</i>! Please see our [Contributing Guide](../CONTRIBUTING.md) for more details on how to get started.
144
+
We welcome contributions to <i>MLflow.js</i>! Please see our [Contributing Guide](../CONTRIBUTING.md) for more details on how to get started.
137
145
138
146
<br>
139
147
@@ -151,4 +159,4 @@ We welcome contributions to <i>mlflow.js</i>! Please see our [Contributing Guide
| Stephany Ho |[](https://github.com/seneyu)|[](https://www.linkedin.com/in/stephanyho/)|
Copy file name to clipboardExpand all lines: mlflow/README.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,6 +1,6 @@
1
1
<!-- README for NPM; the one for GitHub is in .github directory. -->
2
2
3
-
# mlflow.js
3
+
# MLflow.js
4
4
5
5
A JavaScript library designed to provide seamless integration with MLflow's REST API. This package offers access to all the essential endpoints for both the MLflow Tracking Server and Model Registry, along with high-level abstraction workflows, enabling efficient management of machine learning experiments and model lifecycle.
6
6
@@ -31,7 +31,7 @@ createExperiment();
31
31
32
32
# Documentation
33
33
34
-
See [https://www.mlflow-js.org/](https://www.mlflow-js.org/)
34
+
See [https://www.mlflow-js.org/documentation](https://www.mlflow-js.org/documentation)
0 commit comments