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_cookbook and pip 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.json and chat_template.json for 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!