Report an issue

train got OutOfMemoryError('CUDA out of memory. Tried to allocate 384.00 MiB (GPU 0; 14.74 GiB total capacity; 14.22 GiB already allocated; 332.12 MiB free; 14.29 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF')

For example, allow json data as follows json_data = { "text": "Rodrigues works at Microsoft Corporation in Los Angeles. He is gonna travel to an event in New York, that will be held at 03:00 PM on Mar 18, 2024.", "labels": [ {"text": "Rodrigues", "label": "person"}, {"text": "Microsoft Corporation", "label": "company"}, ], "tags": [], }

Add Pabbly Connect so we can easily connect and integrate with different applications associated with CRM, Sales, Marketing, Productivity, or any apps.

Add org scheme to control members, permissions and limits

We're planning to enable a task type "AI Agents", to perform browser-based tasks like searching specific information in a website (crawling) and logging into your 3rd party accounts to consolidate any report, and doing actions like sending an e-mail with a report. Vote on this feature request if you're interested.

Example : The ability to train musical model for recognition of music pieces in .wav or mp3 or mp4 and output as music sheet

At the moment, one is forced to upload one at a time.

To support Unicode font

Enable more custom parameters for training. e.g., model hyperparameters n of synthetic data samples choose hardware (GPUs) choose foundation model (BERT, LLAMA, sentence transformers)