Llama-cookbook v0.0.5 Release Notes
This release changes the package name from llama_recipes to llama_cookbook and deprecates the llama_recipes package name.
Highlighted Changes
- Name:
llama_cookbookandpip install llama-cookbook - Update instructions #852
- Update src package #861, #848
What's Changed
- Improve discoverability of 3.2 recipes by @subramen in #684
- fix readme by @wukaixingxp in #679
- fix AutoModel and bump transformers version to 4.45 by @wukaixingxp in #686
- post1 release version bump by @mreso in #687
- Update multi_modal_infer.py by @init27 in #696
- Add recipe for Llama Triaging & Reporting Tool by @subramen in #651
- Updates to accommodate OpenLLM leaderboard v2 tasks and change Meta Llama 3.1 to Llama 3.1 by @wukaixingxp in #639
- Improve model checkpoint saving logic by @lucas-ventura in #691
- Delete cookie by @init27 in #700
- [Fixed] RuntimeError: probability tensor contains either inf, nan or element < 0 by @himanshushukla12 in #704
- chore: update train_utils.py by @eltociear in #690
- Update requirements.txt by @varunfb in #664
- added missing word and corrected spelling by @jnfinitym in #707
- Fix the bug when continue the peft. by @24kMengXin in #717
- Fix link to LLM finetuning overview by @vvolhejn in #719
- fix Colab link in quickstart_peft_finetuning.ipynb by @jgrivolla in #720
- Fix fine-tuning training loss accumulation by @celestinoalan in #725
- quick fix on readmes and deadlinks by @wukaixingxp in #729
- Initial Crusoe examples to 3p_integrations recipes by @ethxnp in #716
- Fix numpy seed in finetuning.py by @patrik-lambert in #728
- Fix/unit test 3.2 by @mreso in #726
- Tool Calling Tutorial and Example by @init27 in #697
- Update wordlist.txt by @init27 in #736
- Support converting fine-tuned llama 3.2 vision model to HF format and then local inference by @wukaixingxp in #737
- Save the
preprocessor_config.jsonandchat_template.jsonfor mllama model after conversion by @wukaixingxp in #741 - Append epoch rather than best val. loss to val_loss by @celestinoalan in #744
- Update wordlist.txt by @init27 in #745
- Notebook llama by @init27 in #739
- Small notes on next steps by @init27 in #746
- Fix minor grammatical errors by @terrchen in #748
- Update hello_llama_cloud.ipynb by @sharmax-vikas in #753
- Added a Gradio UI for multi-modal inferencing using Llama 3.2 Vision/ by @himanshushukla12 in #718
- Update hello_llama_cloud.ipynb by @sharmax-vikas in #754
- Quickstart docs: Fix path to location of dict for custom datasets by @j-klesen in #755
- Update Step-1 PDF-Pre-Processing-Logic.ipynb by @sharmax-vikas in #756
- added a notebook tutorial on llama 3.2 new capabilities by @Monireh2 in #764
- Updated QuickStart README by @Monireh2 in #767
- fix walkthrough.ipynb render by @subramen in #787
- colab links fix by @jeffxtang in #790
- add TogetherNotebookLM recipe by @zainhas in #766
- add freeze_LLM_only option for mllama finetuning by @JimChienTW in #791
- All functionality has been consolidated into a single file for CLI/UI/Checkpointing and Added fix for issue 702 and added code for that as well, added instructions in local_inference README.md by @himanshushukla12 in #757
- fix typo in auto wrap policy by @hiaoxui in #793
- [docs] small typo in eval readme by @jaysonfrancis in #796
- Add llama 3.2 mmlu, math, gpqa evals to meta_eval harness by @aidando73 in #801
- add 5 together llama notebooks by @zainhas in #798
- Typo in Prompt_Engineering_with_Llama_3.ipynb by @connortreacy in #806
- fix link by @IgorKasianenko in #807
- Browser use with Llama 3.2 Vision Quickstart by @miguelg719 in #799
- Adds Llama 3.2 example on Modal with a fun experiment by @erik-dunteman in #706
- Llama Email Agent by @jeffxtang in #805
- moved images from quickstart to doc/img by @Monireh2 in #823
- Typo in prompt_format_utils.py by @TheAschr in #827
- update text2sql demo to using Llama 3.3 by @jeffxtang in #822
- fix/fix few typos by @abassel in #834
- Update wordlist.txt by @init27 in #833
- fix double bos for vision model by @wukaixingxp in #840
- Refactor Llama-Recipes by @init27 in #832
- Add FAQ by @init27 in #842
- Update README.md by @oneofth3m in #843
- [DRAFT] Fix/refactor recipe references by @IgorKasianenko in #846
- update branch by @init27 in #848
- MM-RAG-New-PR by @init27 in #847
- Add E2B AI Analyst by @jamesmurdza in #768
- Fixed all "Open in Colab" absolute paths by @connortreacy in #850
- Cleaned up API KEY placeholder text by @connortreacy in #851
- Fixing of typo in Step-2-Transcript-Writer.ipynb by @Yggdrasill501 in #858
- Fix package naming by @init27 in #852
- fix few typos by @abassel in #853
- Fix: Updated broken documentations links and references by @iamarunbrahma in #856
- Fix typo by @Yggdrasill501 in #859
- Fix package naming by @init27 in #861
New Contributors
- @lucas-ventura made their first contribution in #691
- @himanshushukla12 made their first contribution in #704
- @jnfinitym made their first contribution in #707
- @24kMengXin made their first contribution in #717
- @vvolhejn made their first contribution in #719
- @jgrivolla made their first contribution in #720
- @celestinoalan made their first contribution in #725
- @ethxnp made their first contribution in #716
- @patrik-lambert made their first contribution in #728
- @terrchen made their first contribution in #748
- @sharmax-vikas made their first contribution in #753
- @j-klesen made their first contribution in #755
- @Monireh2 made their first contribution in #764
- @zainhas made their first contribution in #766
- @JimChienTW made their first contribution in #791
- @hiaoxui made their first contribution in #793
- @jaysonfrancis made their first contribution in #796
- @aidando73 made their first contribution in #801
- @connortreacy made their first contribution in #806
- @IgorKasianenko made their first contribution in #807
- @miguelg719 made their first contribution in #799
- @erik-dunteman made their first contribution in #706
- @TheAschr made their first contribution in #827
- @abassel made their first contribution in #834
- @oneofth3m made their first contribution in #843
- @jamesmurdza made their first contribution in #768
- @Yggdrasill501 made their first contribution in #858
- @iamarunbrahma made their first contribution in #856
Full Changelog: v0.0.4...v0.0.5
Thanks to all contributor for their marvelous efforts!