Product User Community

Hello, we're using pygen to generate a Full Data Model and instances. One of our containers can have multiple connections to objects in another container (see the pic below).As a result, definitions are lists of objects. It is then difficult to query data using tuples: bays = dm_client.bay.list( bay_to_line=(space, id), retrieve_connections="identifier", limit=None, ).to_pandas()We can overcome this by reading all bays with retreive_connections=”full” and then querying the data. But we would like to generate a model, where such tricks are not needed and the definitions are not a list. Is it possible? To generate a model we do this workflow:  neat.read.rdf("lines_bays.jsonld")neat.infer(max_number_of_instance=-1)neat.prepare.data_model.cdf_compliant_external_ids()neat.verify()neat.convert("dms")neat.set.data_model_id(("lin", "lines", "v1"))neat.to.cdf.data_model()neat.to.cdf.instances() 

Hello,  We are currently the Pygen (cognite-pygen==1.0.2) to generate an SDK of our data model. While testing a use case where we want to query all Casings of all Onshore Wells we notice that we get different results using the `List` and `Select` methods. # First method of queryingwells = dm_client.well.list(product_line='Onshore', retrieve_connections='identifier',limit=-1)direct_relations = []for w in wells: if w.wellbores: for wb in w.wellbores: wb_ext=wb.external_id inst = dm.DirectRelationReference(wb.space, wb_ext) direct_relations.append(inst)wb_sections = dm_client.wellbore_section.list(wellbore=direct_relations, retrieve_connections='full', limit=-1)casings = wb_sections.casingprint(f'casings: {len(casings)}') # casings: 645# Second method of queryingcasings= dm_client.well.select().product_line.equals('Onshore').wellbores.wellbore_sections.casing.list_casing(limit=-1)print(f'casings: {len(casings)}') # casings: 94Any thing we missing

Is anyone using Cognite as their main timeseries historian? We are always exploring alternatives and would be interested to hear if Cognite has fit this use case for any users. 

As owner of a canvas, the Canvas UI identifies me as expected such as in version history or in comments. When another user makes edits and comments they are not identified and show up as seen in the attached screenshot. Just in case it matters, this is a Rockwell Automation DataMosaix project. Thanks!

After creating a new calculated time series in Charts, how can I replicate the same calculation across multiple similar assets?Let’s say I calculate [ampers / Fluid rate] in one pump; I’m now interesetd on repeating the same calculation across 10 other pumps (let’s assume timeseries have the same name for all pumps, i.e.: PumpX:AMP).Am I obliged to go to SDK and code that? Shouldn’t be an easier way to do that? Thanks!  

Hi,I have a datamodel in CDF and I want to run a GraphQL query through the SDK to retrieve specific data from the model. When I try to run response = client.data_modeling.graphql.query( id = ('sp_watercourse_data_model', 'Watercourse', 'v1'), query = query )I receive the following error message:CogniteGraphQLError: [GraphQLErrorSpec(message=Could not find data model with space=sp_watercourse_data_model, externalId=Watercourse and version=v1, locations=[], extensions={'classification': 'DataFetchingException'})]However, when I list my models usingmodels = client.data_modeling.data_models.list(limit=100).to_pandas()the model is in the dataframe as expected. What causes this problem and how can I fix it? thanks

I had gotten a number of rest extractor pipe lines running the other day and then paused them. I am now trying to resume them and I am getting an Startup Error message “Missing required field session key”. I do not see a field session key referenced anywhere in the documentation on the rest extractor. Are you able to give any additional inform of where I can find this field?

I have created a function that was configured to 5G of memory and 2 CPUs to run huggingface AI model (py311). Deploying the function went fine. However, when running the function it throws this error.  RuntimeError: [enforce fail at alloc_cpu.cpp:118] err == 0. DefaultCPUAllocator: can't allocate memory: you tried to allocate 9437184 bytes. Error code 12 (Cannot allocate memory)  

Are there any documented use cases or papers on integrating MLflow with Cognite, or is it something we need to implement ourselves?For example, if we aim to seamlessly integrate the MLflow UI with Cognite to evaluate and select the top-performing models, we could leverage SQLite, which operates on the local file system (e.g., mlruns.db) and provides a built-in client, sqlite3. However, our preference is to seamlessly integrate it with Cognite.

Hi,I’ve encountered a bit of issue with CDF when using Microsoft Edge. The full screen button doesn’t do anything. In Chrome and Firefox it does as expected and opens the document in full screen mode.Regards,Markus

Hi, i have set notification to alert when runs failed. However, i noticed that am still receiving email notifications wherever the extraction pipeline is successfully executed. Could you please advice. thanks  

charts Daily average of the previous dayIf anyone knows a good way to do this, please let me know.I would like to see the previous day's 24-hour values by averaging the data over the day with a mean of 0 days 24:00:00 (resampling to granularity), but due to the UTC+9:00 specification, the previous day's processing does not occur until 9:00 AM in japan.I would like to see the previous day's nightly processing volume average at 8:00 AM.The only way I can think of is to use the “shift time series” to force it forward, i would be appriciated if anyone have a better idea.. 

I would like to know the benefitS of implementing UNS architecture in CDF. A detailed guide would be excellent to show how to implement it in the oil and gas refinery with all the prerequisites and third-party platforms along with that.

Hii Team,I'm looking solution for below problem.Currently we have build our custom model using core data model. we have timeseries available in cdf.timeseries resource. Now i want to provide timeseries reference in below view using transformation. Can anyone suggest how timeseries concept is worked in core data model using transformation. type TimeseriesProperty implements Property & CogniteTimeSeries & CogniteDescribable & CogniteSourceable  @view(version: "2_1") {  name: String  source: CogniteSourceSystem  entity: Entity  sourceId: String  sourceContext: String  sourceCreatedTime: Timestamp  sourceUpdatedTime: Timestamp  sourceCreatedUser: String  sourceUpdatedUser: String  isStep: Boolean!  type: CogniteTimeSeries_type  sourceUnit: String  unit: CogniteUnit  assets: [CogniteAsset]  equipment: [CogniteEquipment]  activities: [CogniteActivity]  acquisitionPeriodTimeZone: String  isAggregated: Boolean!  isAggregatedAtSource: Boolean  aggregationMethod: String  isString: Bo

I have set up the Cognite Pi Extractor on a virtual machine and am running into errors in the log:  Fatal error in request of type "CreateTimeSeries": Unauthorized. Unauthorized RequestId: , CDF Message: UnauthorizedI have checked that the client_id and client_secret being used in the config.yml have the ‘time_series: write’ capability for this project. Additionally, I have confirmed that the client_id and client_secret can be used with the python sdk to create timeseries in the project I am working with.Any ideas where I am going wrong?

I’ve had the OPC-UA extractor running for a couple of months with no issues. On Feb 19 it stopped pushing data to Cognite properly. According to Charts, it started pushing again on March 6. As I write this it is March 3rd. The data goes out thru June 2025,The host machine w/extractor has been restarted.The OS date/time looks correct. The log file has the correct date/time. I’d appreciate some advice.  

How is Agentic AI used in Cognite. Are there any use cases where Agentic AL is being used in the oil and gas industry?

Hi,we have quite a number of timeseries which have externalIds with spaces. Unfortunately this throws off the search algorithm of the data explorer (and also in charts when adding new data). Disallowing fuzzy search (which should not be necessary since we *know* the externalid we are looking for) yields no results. Why allowing fuzzy search for the name will give back almost all timeseries not even remotely related to the timeseries we are looking for. I tried placing quotes around the externalId without any success. Is there any way to protect the spaces in the name or to get the search to match externalIds including the spaces?

Can we use azure document intelligence in CDF documents and if I have CDF access then do azure services are free or not as AAD is assigned while creating account? 

Hello team, We have tried to query Entity view which has a property properties of type view Property which is a reverse direct relation.We are using the query endpoint and the instances.query() sdk method to do so.We want the details of properties to come in the select object of Entity. Providing the query below:{ "with": { "0": { "limit": 50, "nodes": { "filter": { "and": [ { "matchAll": {} }, { "hasData": [ { "type": "view", "space": "slb-pdm-dm-governed", "externalId": "Entity", "version": "1_7" } ] }, { "or": [ { "equals": { "property": [

​​​​​​​Can Unified namespace architecture be implemented in Cognite Data Fusion

Hi, I need to add the capability for read/write Timeseries Subscriptions. We are using CI/CD via the Cognite Toolkit. There is no documentation of the different capabilities for this, so I am unsure which Acl to use. I have tried “timeseriesSubscriptionsAcl” and it did not work. What is the correct Acl for Timeseries Subscriptions? Where can I find this documentation? If documentation for this does not exist, can it be created?

Hi Team,I wanted to know if it is possible to integrate Spotfire with Cognite Data Fusion. 

Lets say I have a view called ViewA, which implements CogniteSchedulable.As we know, CogniteSchedulable has the following properties:  Now, according to my requirement, I need to put a default value to the startTime and I also want to make endTime non-nullable. How do I override these settings? Since I am not managing the container definition for these containers, where do I make the changes?Any examples would be appreciated. 

I was checking if i can use the query endpoint (https://api-docs.cognite.com/20230101/tag/Instances/operation/queryContent/) to query data in view using filters on a native timeseries property of that view. in both query explorer as well as data management i observed as such there was no filters supported for native timeseries type property hence wanted to confirm