The Interdecadal Pacific Oscillation
October 28, 2026 at 8:07 PM
The Interdecadal Pacific Oscillation
Time series cross-validation using `crossvalidation` (Part 2)
https://thierrymoudiki.github.io/blog/2021/11/07/r/misc/timeseries/crossvalidation-eval-ts
#Python #DataScience #MachineLearning #rstats #Techtonique
https://thierrymoudiki.github.io/blog/2021/11/07/r/misc/timeseries/crossvalidation-eval-ts
#Python #DataScience #MachineLearning #rstats #Techtonique
October 27, 2026 at 8:19 PM
Time series cross-validation using `crossvalidation` (Part 2)
https://thierrymoudiki.github.io/blog/2021/11/07/r/misc/timeseries/crossvalidation-eval-ts
#Python #DataScience #MachineLearning #rstats #Techtonique
https://thierrymoudiki.github.io/blog/2021/11/07/r/misc/timeseries/crossvalidation-eval-ts
#Python #DataScience #MachineLearning #rstats #Techtonique
#TemporalAggregation is when you reduce the number of data points of a timeseries, increasing the interval between subsequent ones. That is: you make the timeseries less "dense".
October 27, 2026 at 9:20 AM
#TemporalAggregation is when you reduce the number of data points of a timeseries, increasing the interval between subsequent ones. That is: you make the timeseries less "dense".
September data for HadCRUT global temperature now available. 2026 (based on Jan-Sep) best estimate is second warmest in this record, but uncertainties overlap with all of 2023-2025 to some extent.
October 26, 2026 at 11:47 AM
September data for HadCRUT global temperature now available. 2026 (based on Jan-Sep) best estimate is second warmest in this record, but uncertainties overlap with all of 2023-2025 to some extent.
The distinction between the last 4 years comes mostly from differences in their Northern Hemisphere temperature anomalies. Their Southern Hemisphere anomalies are almost the same for the last 4 years.
October 26, 2026 at 11:47 AM
The distinction between the last 4 years comes mostly from differences in their Northern Hemisphere temperature anomalies. Their Southern Hemisphere anomalies are almost the same for the last 4 years.
LSTM Prediction with Recurrent Neural Networks. #BigData #Analytics #DataScience #AI #MachineLearning #IoT #IIoT #Python #RStats #TensorFlow #JavaScript #ReactJS #CloudComputing #Serverless #DataScientist #Linux #Programming #100DaysofCode
geni.us/Timeseries
geni.us/Timeseries
October 26, 2026 at 3:42 AM
LSTM Prediction with Recurrent Neural Networks. #BigData #Analytics #DataScience #AI #MachineLearning #IoT #IIoT #Python #RStats #TensorFlow #JavaScript #ReactJS #CloudComputing #Serverless #DataScientist #Linux #Programming #100DaysofCode
geni.us/Timeseries
geni.us/Timeseries
The identity of a timeseries is its set of attributes and their values, so removing attributes merges timeseries. That leaves several data points at more or less the same timestamp, which need to get merged into one.
October 27, 2026 at 9:20 AM
The identity of a timeseries is its set of attributes and their values, so removing attributes merges timeseries. That leaves several data points at more or less the same timestamp, which need to get merged into one.
There are two fundamental types of timeseries aggregation in #OpenTelemetry: temporal and spatial.
October 27, 2026 at 9:20 AM
There are two fundamental types of timeseries aggregation in #OpenTelemetry: temporal and spatial.
Spatial aggregation is making a metric "narrower" with fewer timeseries active at the same time.
October 27, 2026 at 9:20 AM
Spatial aggregation is making a metric "narrower" with fewer timeseries active at the same time.
It's a big day of the year for climate nerds who follow AMOC: the RAPID array has posted the newest update of their 20-year timeseries monitoring the strength of the overturning circulation.
The takeaway is still muddled. In 2024, it jumped back up in strength, after several years of weakening.
The takeaway is still muddled. In 2024, it jumped back up in strength, after several years of weakening.
Data | Rapid
rapid.ac.uk
October 23, 2026 at 2:32 PM
It's a big day of the year for climate nerds who follow AMOC: the RAPID array has posted the newest update of their 20-year timeseries monitoring the strength of the overturning circulation.
The takeaway is still muddled. In 2024, it jumped back up in strength, after several years of weakening.
The takeaway is still muddled. In 2024, it jumped back up in strength, after several years of weakening.
The latest update for #VictoriaMetrics includes "Finding Why Payments Fail with VictoriaLogs Drilldown" and "#Observability Lessons NVIDIA and CoreWeave".
#DevOps #TimeSeries #OpenSource https://opsmtrs.com/3Lk7JnI
#DevOps #TimeSeries #OpenSource https://opsmtrs.com/3Lk7JnI
VictoriaMetrics
VictoriaMetrics is a fast, cost-effective and scalable open source monitoring solution and time series database typically used for processing high volumes of data and for long term data storage.
opsmtrs.com
October 28, 2026 at 6:16 AM
The latest update for #VictoriaMetrics includes "Finding Why Payments Fail with VictoriaLogs Drilldown" and "#Observability Lessons NVIDIA and CoreWeave".
#DevOps #TimeSeries #OpenSource https://opsmtrs.com/3Lk7JnI
#DevOps #TimeSeries #OpenSource https://opsmtrs.com/3Lk7JnI
Let's talk about timeseries #aggregation: the fine art of starting with a smorgasbord of data points, and ending up with fewer. 🧵
October 27, 2026 at 9:20 AM
Let's talk about timeseries #aggregation: the fine art of starting with a smorgasbord of data points, and ending up with fewer. 🧵
#SpatialAggregation is about merging timeseries: usually you want to throw away some attributes (a.k.a. "labels" in Prometheus) on the data points to reduce #cardinality.
October 27, 2026 at 9:20 AM
#SpatialAggregation is about merging timeseries: usually you want to throw away some attributes (a.k.a. "labels" in Prometheus) on the data points to reduce #cardinality.
IBM Research's Granite Timeseries Ensemble combines four open forecasting models into one tool, with no retraining needed. Permissively licensed on Hugging Face. Run your forecasting workloads at https://massedcompute.com
October 28, 2026 at 12:06 AM
IBM Research's Granite Timeseries Ensemble combines four open forecasting models into one tool, with no retraining needed. Permissively licensed on Hugging Face. Run your forecasting workloads at https://massedcompute.com
#TimeSeries databases face the significant challenge of processing vast amounts of data. At VictoriaMetrics, we are actively developing an #OpenSource Time Series database entirely from scratch using #Go.
October 23, 2026 at 3:39 PM
#TimeSeries databases face the significant challenge of processing vast amounts of data. At VictoriaMetrics, we are actively developing an #OpenSource Time Series database entirely from scratch using #Go.
Next up: Moirai inside agents that reason over time-series data, explain root causes, and recommend actions.
📄 Papers: arxiv.org/abs/2402.02592 (Moirai) and arxiv.org/abs/2511.11698 (Moirai 2.0)
#TimeSeries #FoundationModels #Forecasting #AIResearch
📄 Papers: arxiv.org/abs/2402.02592 (Moirai) and arxiv.org/abs/2511.11698 (Moirai 2.0)
#TimeSeries #FoundationModels #Forecasting #AIResearch
October 28, 2026 at 12:20 PM
Next up: Moirai inside agents that reason over time-series data, explain root causes, and recommend actions.
📄 Papers: arxiv.org/abs/2402.02592 (Moirai) and arxiv.org/abs/2511.11698 (Moirai 2.0)
#TimeSeries #FoundationModels #Forecasting #AIResearch
📄 Papers: arxiv.org/abs/2402.02592 (Moirai) and arxiv.org/abs/2511.11698 (Moirai 2.0)
#TimeSeries #FoundationModels #Forecasting #AIResearch
Ensemble forecasting with IBM Granite time-series models
https://huggingface.co/blog/ibm-research/granite-timeseries-ensemble-r1
https://huggingface.co/blog/ibm-research/granite-timeseries-ensemble-r1
October 22, 2026 at 8:03 AM
Ensemble forecasting with IBM Granite time-series models
https://huggingface.co/blog/ibm-research/granite-timeseries-ensemble-r1
https://huggingface.co/blog/ibm-research/granite-timeseries-ensemble-r1
Discover how AutoML is turning complex financial time‑series into crystal‑clear forecasts. Faster, smarter, more accurate predictions await! #AI #MachineLearning #TimeSeries #AutoML #Finance 🔗
Discover how AutoML is turning complex financial time‑series into crystal‑clear forecasts. Faster, s
bytejournal.online
October 24, 2026 at 9:19 AM
Discover how AutoML is turning complex financial time‑series into crystal‑clear forecasts. Faster, smarter, more accurate predictions await! #AI #MachineLearning #TimeSeries #AutoML #Finance 🔗
That was me on a PR with a colleague whose code was testing the existing of every channel in a timeseries file...
Reminding colleagues that we validate data at the systems edges and fail early is still an invariant.
Reminding colleagues that we validate data at the systems edges and fail early is still an invariant.
October 27, 2026 at 9:59 PM
That was me on a PR with a colleague whose code was testing the existing of every channel in a timeseries file...
Reminding colleagues that we validate data at the systems edges and fail early is still an invariant.
Reminding colleagues that we validate data at the systems edges and fail early is still an invariant.
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads | Netflix TechBlog - Medium
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads
netflixtechblog.com
October 22, 2026 at 4:13 PM
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads | Netflix TechBlog - Medium
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads | Netflix TechBlog - Medium
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads
netflixtechblog.com
October 22, 2026 at 5:13 PM
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads | Netflix TechBlog - Medium
Every single place where you are checking that a column exists needs to be reviewed. We validate the timeseries data so that we do not need to do this. These checks increase the chaotic nature of the code partly because every check suggest that both branches are possible and valid! That is not true
October 27, 2026 at 9:59 PM
Every single place where you are checking that a column exists needs to be reviewed. We validate the timeseries data so that we do not need to do this. These checks increase the chaotic nature of the code partly because every check suggest that both branches are possible and valid! That is not true
Ian Youth: Time series maps in Map Viewer
Tools in ArcGIS Online for animating vector/tabular data with timeseries values in separate columns.
#NACIS2026
Tools in ArcGIS Online for animating vector/tabular data with timeseries values in separate columns.
#NACIS2026
October 21, 2026 at 7:56 PM
Ian Youth: Time series maps in Map Viewer
Tools in ArcGIS Online for animating vector/tabular data with timeseries values in separate columns.
#NACIS2026
Tools in ArcGIS Online for animating vector/tabular data with timeseries values in separate columns.
#NACIS2026
The chart uses average monthly job gains/losses. Let's do the math.
2026: 8k jobs/month x 9 months so far = 72k jobs
2025: -9k x 12 = -108
(It's really -113k jobs for 2025. Chart rounds -9.4k down to -9k.)
The US has lost 41k manufacturing jobs in Trump's second term. data.bls.gov/timeseries/C...
2026: 8k jobs/month x 9 months so far = 72k jobs
2025: -9k x 12 = -108
(It's really -113k jobs for 2025. Chart rounds -9.4k down to -9k.)
The US has lost 41k manufacturing jobs in Trump's second term. data.bls.gov/timeseries/C...
Bureau of Labor Statistics Data
data.bls.gov
October 22, 2026 at 7:45 PM
The chart uses average monthly job gains/losses. Let's do the math.
2026: 8k jobs/month x 9 months so far = 72k jobs
2025: -9k x 12 = -108
(It's really -113k jobs for 2025. Chart rounds -9.4k down to -9k.)
The US has lost 41k manufacturing jobs in Trump's second term. data.bls.gov/timeseries/C...
2026: 8k jobs/month x 9 months so far = 72k jobs
2025: -9k x 12 = -108
(It's really -113k jobs for 2025. Chart rounds -9.4k down to -9k.)
The US has lost 41k manufacturing jobs in Trump's second term. data.bls.gov/timeseries/C...