Granted Patent B2

US 12,645,693 · App. 17/825,323 · Granted Jun 2, 2026

Inventors: Laurent Parmentier (Marcq-En-Baroeul, FR); Olivier Nicol (Tourcoing, FR); Clement Bataille (Lille, FR)

Assignee: OVH

G06F16/258G06N20/00

Abstract

Claims (51)

1 . A method for sharing a source dataset between a first user device and a second user device, the method being executed by a first processor communicably connected to the first user device, the method comprising:

receiving a request to initiate sharing of the source dataset;

accessing a configuration file comprising a plurality of instruction sets, each instruction set enabling retrieval of one or more source datasets stored in a plurality of respective data storage services;

receiving an indication from a user associated with the first user device, the indication identifying a selected source dataset stored in a corresponding selected data storage service from the respective data storage services, the indication including a metadata requirement, and the indication further identifying a target data format for the selected source dataset;

storing, in the configuration file, settings corresponding to the indication from the user;

extracting, from the configuration file, an instruction set corresponding to the selected source dataset and the settings;

executing the extracted instruction set and the settings, and retrieving the selected source dataset from the selected data storage service onto the first user device, wherein retrieving the selected source dataset comprises selecting the source dataset according to the metadata requirement;

providing the selected source dataset retrieved by the first user device to a first machine learning model for training thereof;

sharing the selected source dataset by transferring the configuration file to the second user device to enable the second user device to, when executing the extracted instruction set, retrieve the selected source dataset from the selected data storage service in the target data format;

extracting, from the configuration file and by a second processor communicably connected to the second user device, the instruction set corresponding to the selected source dataset and the settings;

retrieving the selected source dataset from the selected data storage service onto the second user device by executing the extracted instruction set and the settings, wherein the selected source dataset retrieved by the second user device holds identical data as the selected source dataset retrieved by the first user device; and

providing, by the second processor, the selected source dataset retrieved by the second user device to a second machine learning model for training thereof.

2 . The method of claim 1 , wherein the selected source dataset is a first selected source dataset from a first selected data storage service; and

further comprising:

retrieving, by the first processor, a second selected source dataset from a second selected data storage service onto the first user device;

retrieving, by the second processor, the second selected source dataset from a second selected data storage service onto the second user device; and

providing both of the first and second machine learning models with the second selected source dataset for training thereof.

3 . The method of claim 1 , wherein:

the instruction set comprises instructions to, when executed, convert the selected source dataset, from a first data format into the target data format.

4 . The method of claim 3 , wherein converting the selected source dataset comprises at least one of:

converting the selected source dataset into a desired file type; and

converting the selected source dataset into a desired data formatting style.

5 . The method of claim 1 , wherein:

the indication further includes a data requirement; and

retrieving the selected source dataset from the selected data storage service comprises selecting the dataset according to the data requirement.

6 . A system comprising a first user device and a second user device, wherein the first user device comprises at least one processor and memory comprising executable instructions which, when executed by the at least one processor of the first user device, cause the first user device to:

receive a request to initiate sharing of a source dataset;

access a configuration file comprising a plurality of instruction sets, each instruction set enabling retrieval of one or more source datasets stored in a plurality of respective data storage services;

receive an indication from a user associated with the first user device, the indication identifying a selected source dataset stored in a corresponding selected data storage service from the respective data storage services, the indication including a metadata requirement, and the indication further identifying a target data format for the selected source dataset;

store, in the configuration file, settings corresponding to the indication from the user;

extract, from the configuration file, an instruction set corresponding to the selected source dataset and the settings;

retrieve the selected source dataset from the selected data storage service by executing the extracted instruction set and the settings, wherein retrieving the selected source dataset comprises selecting the source dataset according to the metadata requirement;

provide the selected source dataset retrieved by the first user device to a first machine learning model for training thereof; and

send the configuration file to the second user device, and

wherein the second user device comprises at least one processor and memory comprising executable instructions which, when executed by the at least one processor of the second user device, cause the second user device to:

receive the configuration file;

extract, from the configuration file, the instruction set corresponding to the selected source dataset and the settings;

execute the extracted instruction set;

retrieve the selected source dataset from the selected data storage service in the target data format by executing the extracted instruction set and the settings, wherein the selected source dataset retrieved by the second user device holds identical data as the selected source dataset retrieved by the first user device; and

provide the selected source dataset retrieved by the second user device to a second machine learning model for training thereof.

7 . The system of claim 6 , wherein:

the instruction set comprises instructions to, when executed, convert the selected source dataset, from a first data format into the target data format.

8 . The system of claim 7 , wherein converting the selected source dataset comprises at least one of:

converting the selected source dataset into a desired file type; and

converting the selected source dataset into a desired data formatting style.

9 . The system of claim 6 , wherein:

the indication further includes a data requirement; and

wherein the instructions that cause the first user device to retrieve the selected source dataset from the selected data storage service comprise instructions that cause the first user device to select the dataset according to the data requirement.

10 . The system of claim 6 , wherein:

the indication further includes a metadata requirement; and

wherein the instructions that cause the first user device to retrieve the selected source dataset from the selected data storage service comprise instructions that cause the first user device to select the dataset according to the metadata requirement.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (139)