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
Copy file name to clipboardExpand all lines: README.md
+9-3Lines changed: 9 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -56,7 +56,7 @@ SELECT * FROM pgml.train(
56
56
'Human-friendly project name',
57
57
'regression',
58
58
'<name of the table or view containing the data>',
59
-
'<name of the column containing the y target value>'
59
+
'<name of the column containing the y target values>'
60
60
);
61
61
```
62
62
@@ -70,7 +70,7 @@ Once the model is trained, making predictions is as simple as:
70
70
SELECTpgml.predict('Human-friendly project name', ARRAY[...]) AS prediction_score;
71
71
```
72
72
73
-
where `ARRAY[...]` is a list of the features for which we want to run a prediction. This list has to be in the same order as the columns in the data table. This score then can be used in normal queries, for example:
73
+
where `ARRAY[...]` is a list of features for which we want to run a prediction. This list has to be in the same order as the columns in the data table. This score then can be used in normal queries, for example:
74
74
75
75
```sql
76
76
SELECT*,
@@ -92,7 +92,7 @@ As data in your database changes, it is possible to retrain the model again to g
92
92
93
93
## Roadmap
94
94
95
-
This project is currently a proof of concept. Some important features which we are currently thinking about or working on are listed below.
95
+
This project is currently a proof of concept. Some important features, which we are currently thinking about or working on, are listed below.
96
96
97
97
### Production deployment
98
98
@@ -108,11 +108,17 @@ A good looking and useful UI goes a long way. A dashboard similar to existing so
108
108
109
109
A data explorer allows anyone to browse the dataset in production and to find useful tables and features to build effective machine learning models.
110
110
111
+
111
112
### More algorithms
112
113
113
114
Scikit-Learn is a good start, but we're also thinking about including Tensorflow, Pytorch, and many more useful models.
114
115
115
116
117
+
### Scheduled training
118
+
119
+
In applications where data changes often, it's useful to retrain the models automatically on a schedule, e.g. every day, every week, etc.
0 commit comments