@@ -25,7 +25,7 @@ pip install faim-sdk
2525
2626## Authentication
2727
28- Get your free API key at ** [ https://faim.it.com/ ] ( https://faim.it.com/ ) **
28+ Get your API key at ** [ https://faim.it.com/ ] ( https://faim.it.com/ ) **
2929
3030``` python
3131from faim_sdk import ForecastClient
@@ -459,81 +459,11 @@ async def forecast_multiple_series():
459459responses = asyncio.run(forecast_multiple_series())
460460```
461461
462- ## Configuration
463-
464- ### Client Options
465-
466- ``` python
467- from faim_sdk import ForecastClient
468-
469- # Basic configuration
470- client = ForecastClient(
471- timeout = 120.0 , # Request timeout in seconds (default: 120)
472- verify_ssl = True , # SSL certificate verification (default: True)
473- )
474-
475- # With API key authentication
476- client = ForecastClient(
477- api_key = " your-secret-api-key" ,
478- timeout = 120.0
479- )
480-
481- # Advanced configuration with custom httpx settings
482- import httpx
483-
484- client = ForecastClient(
485- api_key = " your-api-key" ,
486- timeout = 120.0 ,
487- limits = httpx.Limits(max_connections = 10 ), # Connection pooling
488- headers = {" X-Custom-Header" : " value" } # Custom headers
489- )
490- ```
491-
492- ### Request Options
493-
494- ``` python
495- # Compression options for large payloads
496- request = Chronos2ForecastRequest(
497- x = data,
498- horizon = 24 ,
499- compression = " zstd" # Options: "zstd", "lz4", None (default: "zstd")
500- )
501-
502- # Model version pinning
503- request = FlowStateForecastRequest(
504- x = data,
505- horizon = 24 ,
506- model_version = " 1.2.3" # Pin to specific version (default: "latest")
507- )
508- ```
509-
510- ## Context Managers
511-
512- Use context managers for automatic resource cleanup:
513-
514- ``` python
515- # Sync context manager
516- with ForecastClient() as client:
517- request = Chronos2ForecastRequest(x = data, horizon = 24 , quantiles = [0.1 , 0.5 , 0.9 ])
518- response = client.forecast(request)
519- print (response.quantiles)
520- # Client automatically closed
521-
522- # Async context manager
523- async with ForecastClient() as client:
524- request = Chronos2ForecastRequest(x = data, horizon = 24 , quantiles = [0.1 , 0.9 ])
525- response = await client.forecast_async(request)
526- print (response.quantiles)
527- # Client automatically closed
528- ```
529-
530462## Examples
531463
532464See the ` examples/ ` directory for complete Jupyter notebook examples:
533465
534- - ** ` flowstate_simple_example.ipynb ` ** - Point forecasting with FlowState on AirPassengers dataset
535- - ** ` chronos2_probabilistic.ipynb ` ** - Probabilistic forecasting with quantiles (coming soon)
536- - ** ` batch_processing.ipynb ` ** - Efficient batch processing patterns (coming soon)
466+ - ** ` model_comparison_simple.ipynb ` ** - Point forecasting with FlowState on AirPassengers dataset
537467
538468## Requirements
539469
@@ -562,9 +492,7 @@ See the `examples/` directory for complete Jupyter notebook examples:
562492
563493## Support
564494
565- - ** Documentation** : [ docs.faim.example.com] ( https://docs.faim.example.com )
566- - ** Issues** : [ GitHub Issues] ( https://github.com/your-org/faim-sdk/issues )
567- - ** Email** : support@faim.example.com
495+ - ** Email** : support@faim.it.com
568496
569497## License
570498
@@ -579,6 +507,6 @@ If you use FAIM in your research, please cite:
579507 title = {FAIM SDK: Foundation AI Models for Time Series Forecasting},
580508 author = {FAIM Team},
581509 year = {2024},
582- url = {https://github.com/your-org /faim-sdk }
510+ url = {https://github.com/S-FM /faim-python-client }
583511}
584512```
0 commit comments