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
15
16
-
<ahref="">Visit the official mlflow.js site for more info!</a>
16
+
<ahref="https://www.mlflow-js.org/">Visit the official <i>mlflow.js</i> site for more info!</a>
<i>mlflow.js</i> covers all REST API endpoints under MLflow's Tracking Server and Model Registry. Official documentation for MLflow.js can be found <ahref="">here</a>. Moreover, high-level abstraction workflows have been developed to facilitate developers' work processes.
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
27
28
28
### High-Level Abstraction Workflows
29
29
30
30
**Experiment Manager**
31
31
32
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, ogging mettrics, params,tags, and the model, and finishing the run
34
-
- experimentSummary - Returns an array of all the passed-in experiment's runs, sorted accoroding to the passed-in metric
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
35
35
36
36
**Run Manager**
37
37
@@ -40,9 +40,15 @@ Visit our LinkedIn page below:
40
40
41
41
**Model Manager**
42
42
43
-
- createRegisteredModelWithVersion - Creates a new registered model and the frist version of that model
43
+
- createRegisteredModelWithVersion - Creates a new registered model and the first version of that model
44
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 foro tthat latest version
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
46
52
47
53
<br>
48
54
@@ -73,7 +79,7 @@ Ensure MLflow is installed on your system:
73
79
pip install mlflow
74
80
```
75
81
76
-
Note: MLflow is compatible with MacOS. If you encountner issues with the default system Python, consider installing Python 3 via the Homebrew package manger using `brew install python`. In this case, installing MLflow is now `pip3 install mlflow`.
82
+
**Note:** MLflow is compatible with MacOS. If you encounter issues with the default system Python, consider installing Python 3 via the Homebrew package manger using `brew install python`. In this case, installing MLflow is now `pip3 install mlflow`.
77
83
78
84
### Start the MLflow Tracking Server
79
85
@@ -89,66 +95,60 @@ This will launch the MLflow UI on your local machine at `http://localhost:5000`.
89
95
90
96
## Quickstart
91
97
92
-
### Install MLflow.js Library
98
+
### Install <i>mlflow.js</i> Library
93
99
94
-
To use the MLflow.js library, navigate to your project directory and install it via npm:
100
+
To use the <i>mlflow.js</i> library, navigate to your project directory and install it via npm:
95
101
96
102
```bash
97
103
npm install mlflow-js
98
104
```
99
105
100
106
### Usage Example
101
107
102
-
Here is an example of how to use the MLflow.js library to create an experiment:
108
+
Here is an example of how to use the <i>mlflow.js</i> library to create an experiment:
|Stephany Ho |[](https://github.com/seneyu)|[](https://www.linkedin.com/in/stephanyho/)|
|Stephany Ho |[](https://github.com/seneyu)|[](https://www.linkedin.com/in/stephanyho/)|
Copy file name to clipboardExpand all lines: mlflow-site/src/app/components/Demo.tsx
+6-8Lines changed: 6 additions & 8 deletions
Original file line number
Diff line number
Diff line change
@@ -3,19 +3,17 @@ import DemoCard from "./DemoCard";
3
3
constDemo=()=>{
4
4
constdemos=[];
5
5
constdemoCardHeaders=[
6
-
'Demo 1',
7
-
'Demo 2',
8
-
'Demo 3'
6
+
'Manage experiments',
7
+
'Complete workflow',
9
8
];
10
9
constdemoCardBlurbs=[
11
-
'Description for what\'s happening in Demo 1. This block of text will contain all the info needed to understand the demo.',
12
-
'Description for what\'s happening in Demo 2. This block of text will contain all the info needed to understand the demo.',
13
-
'Description for what\'s happening in Demo 3. This block of text will contain all the info needed to understand the demo.'
10
+
'Create experiments with MLflow.js. Using built-in workflows, manage complex operations easily.',
11
+
'Use MLflow.js to support a full ML project with TensorFlow.js. Log hyperparameters and key metrics during each training step. Evaluate model performance and register succesful models.',
Copy file name to clipboardExpand all lines: mlflow-site/src/app/components/Features.tsx
+9-9Lines changed: 9 additions & 9 deletions
Original file line number
Diff line number
Diff line change
@@ -3,20 +3,20 @@ import FeatureCard from "./FeatureCard";
3
3
constFeatures=()=>{
4
4
constfeatureHeader='MLOps in Javascript, made simple.';
5
5
constfeatureLongBlurb=`
6
-
Longer blurb about MLFlow.js\'s feature set. Longer blurb about MLFlow.js\'s feature set. Longer blurb about MLFlow.js\'s feature set. Longer blurb about MLFlow.js\'s feature set. Longer blurb about MLFlow.js\'s feature set. Longer blurb about MLFlow.js\'s feature set.
6
+
MLflow.js makes ML experimentation and model management seamless for JavaScript developers. Built with TypeScript, it provides intuitive access to MLflow\'s complete REST API while adding powerful abstractions for common ML workflows. Whether you\'re training models with TensorFlow.js, managing A/B tests, or monitoring production models, MLflow.js helps you track everything in one place.
'Connect your JavaScript stack directly to MLflow with minimal setup.',
17
+
'Automate key MLOps tasks directly from Node.js, simplifying workflow management. Manage experiments, runs, model registry and model version management with dedicated methods.',
18
+
'Designed specifically for JavaScript developers: no Python knowledge required.',
19
+
'Execute complex MLOps tasks with a single function call with MLflow.js\'s powerful built-in workflows.'
0 commit comments