Releases: togethercomputer/together-python
Releases · togethercomputer/together-python
v1.0.1
What's Changed
- [WIP] v1.0.0 Updates by @orangetin in #71
- Bump black from 23.12.1 to 24.3.0 by @dependabot in #76
- remove unused file _utils.py by @orangetin in #74
- Add basic issue templates by @orangetin in #75
- Create dependabot.yml by @orangetin in #77
- Fix annotation compatibility for older python versions by @orangetin in #78
- [WIP] Docs for v1.0.0 by @Nutlope in #84
- Bump pytest from 7.4.4 to 8.1.1 by @dependabot in #80
- Bump filelock from 3.13.1 to 3.13.3 by @dependabot in #79
- Bump typer from 0.9.0 to 0.12.0 by @dependabot in #81
- Bump ruff from 0.3.4 to 0.3.5 by @dependabot in #82
- Bump types-requests from 2.31.0.20240311 to 2.31.0.20240402 by @dependabot in #83
- CLI related fixes and improvements by @orangetin in #85
- Legacy compatibility by @orangetin in #87
- Fix typos in readme for files/fine-tune by @orangetin in #88
New Contributors
- @dependabot made their first contribution in #76
- @Nutlope made their first contribution in #84
Full Changelog: v0.2.11...v1.0.1
Contributors
dependabot, Nutlope, and orangetin
Assets 2
v0.2.11
Contributors
clam004, justusc, and azahed98
Assets 2
v0.2.10
v0.2.9
v0.2.8
What's Changed
- update docs with model_output_name and wandb_url by @clam004 in #58
- Embeddings api by @clam004 in #59
- Add Openai compatible Completion class by @orangetin in #60
- Fix workflow and bump to version 0.2.8 by @orangetin in #61
Full Changelog: v0.2.6...v0.2.8
Contributors
clam004 and orangetin
Assets 2
v0.2.6
What's Changed
- Data feedback rdme by @clam004 in #21
- Fix typo in error message of error code 429 in Complete.py by @rchalamala in #22
- Price estimate for FT jobs by @adamsch1 in #23
- add support for 70b by @adamsch1 in #24
- Enable WandB option in the CLI by @justusc in #25
- Remove 70B models (not yet supported) by @justusc in #26
- version update by @clam004 in #27
- Add 70B models support for fine tuning job submission by @justusc in #28
- Add negative prompt as a param for image models by @orangetin in #31
- Add a minimum for finetune create batch size by @justusc in #32
- version by @clam004 in #33
- together models ready model_name, but using the API instead by @clam004 in #34
- Timestamp formats by @clam004 in #36
- Fix circular imports by @orangetin in #38
- Version bump to 0.2.4 by @orangetin in #39
- Clean up fixed constants available with api call and add new functions by @orangetin in #40
- Confirm Finetune Job by @clam004 in #41
- updates finetune input resets to the October 70b update by @clam004 in #42
- Deduplicate Utils by @clam004 in #43
- Do not show the stack trace when user as no API key by @clam004 in #46
- Add max batch size check by @azahed98 in #47
- Replace ValueError for max batchsize with ajusted_inputs=True by @azahed98 in #48
- typing errors with python <3.9 by @clam004 in #49
- Tabulate outputs for JSON outputs by @orangetin in #50
- Initial tests for files and fine-tuning by @orangetin in #51
- Enable retries for file uploads and finetune downloads by @orangetin in #52
- Handle errors gracefully by @orangetin in #53
- Pypi push on github release creation by @orangetin in #54
- Fix poetry install by @orangetin in #55
- Fix poetry build by @orangetin in #56
New Contributors
- @rchalamala made their first contribution in #22
- @adamsch1 made their first contribution in #23
- @azahed98 made their first contribution in #47
Full Changelog: v0.1.5...v0.2.6
Contributors
adamsch1, clam004, and 4 other contributors
Assets 2
v0.1.5
v0.1.4
v0.1.4 - 2023-08-18
Full Changelog: v0.0.9...v0.1.4
Changes
- Step by step guidance and examples added to documentation for the completion, files, and finetune sections, including example jsonl dataset that can be downloaded locally for inspection, automated checking and uploading
- optional but very useful together.Files.check function to check the correctness of jsonl files. Gives useful feeback on which first line that is formatted incorrectly, will give the specific reason for being the line being incorrect and how to fix it with a helpful link
- model specific token option to check if model specific tokens are used in dataset
- together.Finetune.create will give useful feedback if user tries to upload anything other than a valid file-id for a file they already have uploaded successfully
Here is an example file check output:
{ 'file_present': 'File found',
'file_size': 'File size 0.2 GB',
'is_check_passed': False,
'key_value': 'Unexpected, value type for "text" key on line 6 of the input '
'file.The value type of the "text" key must be a '
'string.Expected format: {"text":"my sample string"}See '
'https://docs.together.ai/docs/fine-tuning for more '
'information.{"text": {"text":"<human>: Salutations!\\n<bot>: '
'Salutations to you as well! How can I help you today?"}}\n',
'min_samples': 'Processing /Users/carsonlam/Projects/data/bad.jsonl '
'resulted in only 10 samples. Our minimum is 100 samples. ',
'model_special_tokens': 'we are not yet checking end of sentence tokens '
'for this model',
'num_samples_w_eos_token': 0,
'text_field': 'No "text" field was found on line 7 of the the input '
'file.Expected format: {"text":"my sample string"}.see '
'https://docs.together.ai/docs/fine-tuning for more '
'information.{"ext": "<human>: Hiya!\\n<bot>: Hiya! How may '
'I assist you?"}\n'}
v0.1.3
v0.1.3 - 2023-08-11
Full Changelog: v0.0.9...v0.1.3
Changes
- Users can now add their weights and biases WANDB_API_KEY
- For fine-tuning, users indicate the number of checkpoints using the
--n-checkpointsparameter
here is an example:
together finetune create -t file-960be810-4d33-449a-885a-9f69bd8fd0e2 -m togethercomputer/LLaMA-2-7B-32K --n-checkpoints 2 --wandb-api-key 1123456789abcdefghijklmnop