Fix more package naming (#864)

3p-integrations/crusoe/vllm-fp8/README.md

@@ -23,8 +23,8 @@ source $HOME/.cargo/env
 
 Now, clone the recipes and navigate to this tutorial. Initialize the virtual environment and install dependencies:
 ```bash
-git clone https://github.com/meta-llama/llama-recipes.git
-cd llama-recipes/recipes/3p_integrations/crusoe/vllm-fp8/
+git clone https://github.com/meta-llama/llama-cookbook.git
+cd llama-cookbook/recipes/3p_integrations/crusoe/vllm-fp8/
 uv add vllm setuptools
 ```
 

3p-integrations/tgi/README.md

@@ -9,7 +9,7 @@ In case the model was fine tuned with LoRA method we need to merge the weights o
 The script takes the base model, the peft weight folder as well as an output as arguments:
 
 ```
-python -m llama_recipes.recipes.3p_integration.tgi.merge_lora_weights --base_model llama-7B --peft_model ft_output --output_dir data/merged_model_output
+python -m llama_cookbook.recipes.3p_integration.tgi.merge_lora_weights --base_model llama-7B --peft_model ft_output --output_dir data/merged_model_output
 ```
 
 ## Step 1: Serving the model

end-to-end-use-cases/benchmarks/llm_eval_harness/meta_eval/README.md

@@ -25,8 +25,8 @@ Given those differences, the numbers from this recipe can not be compared to the
 Please install lm-evaluation-harness and our llama-recipe repo by following:
 
 ```
-git clone git@github.com:meta-llama/llama-recipes.git
-cd llama-recipes
+git clone git@github.com:meta-llama/llama-cookbook.git
+cd llama-cookbook
 pip install -U pip setuptools
 pip install -e .
 pip install lm-eval[math,ifeval,sentencepiece,vllm]==0.4.3

end-to-end-use-cases/coding/text2sql/quickstart.ipynb

@@ -1,5 +1,5 @@
 {
- "cells": [
+ "cells": [llama-cookbook
   {
    "cell_type": "markdown",
    "id": "e8cba0b6",

end-to-end-use-cases/customerservice_chatbots/RAG_chatbot/RAG_Chatbot_Example.ipynb

@@ -402,7 +402,7 @@
     "In this example, we will be deploying a Meta Llama 3 8B chat HuggingFace model with the Text-generation-inference framework on-permises.  \n",
     "This would allow us to directly wire the API server with our chatbot.  \n",
     "There are alternative solutions to deploy Meta Llama 3 models on-permises as your local API server.  \n",
-    "You can find our complete guide [here](https://github.com/meta-llama/llama-recipes/blob/main/recipes/inference/model_servers/llama-on-prem.md)."
+    "You can find our complete guide [here](https://github.com/meta-llama/llama-cookbook/blob/main/recipes/inference/model_servers/llama-on-prem.md)."
    ]
   },
   {

end-to-end-use-cases/github_triage/README.md

@@ -32,7 +32,7 @@ pip install -r requirements.txt
 ### Running the Tool
 
 ```bash
-python triage.py --repo_name='meta-llama/llama-recipes' --start_date='2024-08-14' --end_date='2024-08-27'
+python triage.py --repo_name='meta-llama/llama-cookbook' --start_date='2024-08-14' --end_date='2024-08-27'
 ```
 
 ### Output

end-to-end-use-cases/github_triage/walkthrough.ipynb

@@ -1,4 +1,4 @@
-{
+{llama-cookbookllama-cookbook
   "cells": [
     {
       "cell_type": "code",

end-to-end-use-cases/responsible_ai/code_shield_usage_demo.ipynb

@@ -151,7 +151,7 @@
     "import os\n",
     "import getpass\n",
     "\n",
-    "from llama_recipes.inference.llm import TOGETHER, OPENAI, ANYSCALE\n",
+    "from llama_cookbook.inference.llm import TOGETHER, OPENAI, ANYSCALE\n",
     "\n",
     "if \"EXTERNALLY_HOSTED_LLM_TOKEN\" not in os.environ:\n",
     "    os.environ[\"EXTERNALLY_HOSTED_LLM_TOKEN\"] = getpass.getpass(prompt=\"Provide token for LLM provider\")\n",

end-to-end-use-cases/responsible_ai/llama_guard/README.md

@@ -6,7 +6,7 @@ This [notebook](llama_guard_text_and_vision_inference.ipynb) shows how to load t
 
 ## Requirements
 1. Access to Llama guard model weights on Hugging Face. To get access, follow the steps described in the top of the model card in [Hugging Face](https://huggingface.co/meta-llama/Llama-Guard-3-1B)
-2. Llama recipes package and its dependencies [installed](https://github.com/meta-llama/llama-recipes?tab=readme-ov-file#installing)
+2. Llama recipes package and its dependencies [installed](https://github.com/meta-llama/llama-cookbook?tab=readme-ov-file#installing)
 3. Pillow package installed
 
 ## Inference Safety Checker

end-to-end-use-cases/responsible_ai/llama_guard/llama_guard_customization_via_prompting_and_fine_tuning.ipynb

@@ -33,7 +33,7 @@
     "\n",
     "Llama Guard is provided with a reference taxonomy explained on [this page](https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-guard-3), where the prompting format is also explained. \n",
     "\n",
-    "The functions below combine already existing [prompt formatting code in llama-recipes](https://github.com/meta-llama/llama-recipes/blob/main/src/llama_recipes/inference/prompt_format_utils.py) with custom code to aid in the custimization of the taxonomy. "
+    "The functions below combine already existing [prompt formatting code in llama-recipes](https://github.com/meta-llama/llama-recipes/blob/main/src/llama_cookbook/inference/prompt_format_utils.py) with custom code to aid in the custimization of the taxonomy. "
    ]
   },
   {
@@ -80,7 +80,7 @@
    ],
    "source": [
     "from enum import Enum\n",
-    "from llama_recipes.inference.prompt_format_utils import  LLAMA_GUARD_3_CATEGORY, SafetyCategory, AgentType\n",
+    "from llama_cookbook.inference.prompt_format_utils import  LLAMA_GUARD_3_CATEGORY, SafetyCategory, AgentType\n",
     "from typing import List\n",
     "\n",
     "class LG3Cat(Enum):\n",
@@ -158,7 +158,7 @@
     }
    ],
    "source": [
-    "from llama_recipes.inference.prompt_format_utils import build_custom_prompt, create_conversation, PROMPT_TEMPLATE_3, LLAMA_GUARD_3_CATEGORY_SHORT_NAME_PREFIX\n",
+    "from llama_cookbook.inference.prompt_format_utils import build_custom_prompt, create_conversation, PROMPT_TEMPLATE_3, LLAMA_GUARD_3_CATEGORY_SHORT_NAME_PREFIX\n",
     "from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig\n",
     "from typing import List, Tuple\n",
     "from enum import Enum\n",
@@ -463,13 +463,13 @@
     "\n",
     "To add additional datasets\n",
     "\n",
-    "1. Copy llama-recipes/src/llama_recipes/datasets/toxicchat_dataset.py \n",
+    "1. Copy llama-recipes/src/llama_cookbook/datasets/toxicchat_dataset.py \n",
     "2. Modify the file to change the dataset used\n",
     "3. Add references to the new dataset in \n",
-    "    - llama-recipes/src/llama_recipes/configs/datasets.py\n",
-    "    - llama_recipes/datasets/__init__.py\n",
-    "    - llama_recipes/datasets/toxicchat_dataset.py\n",
-    "    - llama_recipes/utils/dataset_utils.py\n",
+    "    - llama-recipes/src/llama_cookbook/configs/datasets.py\n",
+    "    - llama_cookbook/datasets/__init__.py\n",
+    "    - llama_cookbook/datasets/toxicchat_dataset.py\n",
+    "    - llama_cookbook/utils/dataset_utils.py\n",
     "\n",
     "\n",
     "## Evaluation\n",
@@ -484,7 +484,7 @@
    "source": [
     "from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig\n",
     "\n",
-    "from llama_recipes.inference.prompt_format_utils import build_default_prompt, create_conversation, LlamaGuardVersion\n",
+    "from llama_cookbook.inference.prompt_format_utils import build_default_prompt, create_conversation, LlamaGuardVersion\n",
     "from llama.llama.generation import Llama\n",
     "\n",
     "from typing import List, Optional, Tuple, Dict\n",
@@ -726,7 +726,7 @@
     "#     \"unsafe_content\": [\"O1\"]\n",
     "# }\n",
     "# ```\n",
-    "from llama_recipes.datasets.toxicchat_dataset import get_llamaguard_toxicchat_dataset\n",
+    "from llama_cookbook.datasets.toxicchat_dataset import get_llamaguard_toxicchat_dataset\n",
     "validation_data = get_llamaguard_toxicchat_dataset(None, None, \"train\", return_jsonl = True)[0:100]\n",
     "run_validation(validation_data, AgentType.USER, Type.HF, load_in_8bit = False, load_in_4bit = True)"
    ]
@@ -757,7 +757,7 @@
    "outputs": [],
    "source": [
     "model_id = \"meta-llama/Llama-Guard-3-8B\"\n",
-    "from llama_recipes import finetuning\n",
+    "from llama_cookbook import finetuning\n",
     "\n",
     "finetuning.main(\n",
     "    model_name = model_id,\n",

end-to-end-use-cases/responsible_ai/prompt_guard/README.md

@@ -8,4 +8,4 @@ This is a very small model and inference and fine-tuning are feasible on local C
 
 ## Requirements
 1. Access to Prompt Guard model weights on Hugging Face. To get access, follow the steps described [here](https://github.com/facebookresearch/PurpleLlama/tree/main/Prompt-Guard#download)
-2. Llama recipes package and it's dependencies [installed](https://github.com/meta-llama/llama-recipes?tab=readme-ov-file#installing)
+2. Llama recipes package and it's dependencies [installed](https://github.com/meta-llama/llama-cookbook?tab=readme-ov-file#installing)

getting-started/README.md

@@ -1,4 +1,4 @@
-## Llama-Recipes Getting Started
+## Llama-cookbook Getting Started
 
 If you are new to developing with Meta Llama models, this is where you should start. This folder contains introductory-level notebooks across different techniques relating to Meta Llama.
 

getting-started/finetuning/finetune_vision_model.md

@@ -1,7 +1,7 @@
 ## Llama 3.2 Vision Models Fine-Tuning Recipe
 This recipe steps you through how to finetune a Llama 3.2 vision model on the OCR VQA task using the [OCRVQA](https://huggingface.co/datasets/HuggingFaceM4/the_cauldron/viewer/ocrvqa?row=0) dataset.
 
-**Disclaimer**: As our vision models already have a very good OCR ability, here we use the OCRVQA dataset only for demonstration purposes of the required steps for fine-tuning our vision models with llama-recipes.
+**Disclaimer**: As our vision models already have a very good OCR ability, here we use the OCRVQA dataset only for demonstration purposes of the required steps for fine-tuning our vision models with llama-cookbook.
 
 ### Fine-tuning steps
 

getting-started/finetuning/finetuning.py

@@ -2,7 +2,7 @@
 # This software may be used and distributed according to the terms of the Llama 2 Community License Agreement.
 
 import fire
-from llama_recipes.finetuning import main
+from llama_cookbook.finetuning import main
 
 if __name__ == "__main__":
     fire.Fire(main)

getting-started/finetuning/quickstart_peft_finetuning.ipynb

@@ -31,17 +31,17 @@
    "source": [
     "### Step 0: Install pre-requirements and convert checkpoint\n",
     "\n",
-    "We need to have llama-recipes and its dependencies installed for this notebook. Additionally, we need to log in with the huggingface_cli and make sure that the account is able to to access the Meta Llama weights."
+    "We need to have llama-cookbook and its dependencies installed for this notebook. Additionally, we need to log in with the huggingface_cli and make sure that the account is able to to access the Meta Llama weights."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "# uncomment if running from Colab T4\n",
-    "# ! pip install llama-recipes ipywidgets\n",
+    "# ! pip install llama-cookbook ipywidgets\n",
     "\n",
     "# import huggingface_hub\n",
     "# huggingface_hub.login()"
@@ -59,7 +59,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
    "outputs": [
     {
@@ -80,7 +80,7 @@
    "source": [
     "import torch\n",
     "from transformers import LlamaForCausalLM, AutoTokenizer\n",
-    "from llama_recipes.configs import train_config as TRAIN_CONFIG\n",
+    "from llama_cookbook.configs import train_config as TRAIN_CONFIG\n",
     "\n",
     "train_config = TRAIN_CONFIG()\n",
     "train_config.model_name = \"meta-llama/Meta-Llama-3.1-8B\"\n",
@@ -221,8 +221,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from llama_recipes.configs.datasets import samsum_dataset\n",
-    "from llama_recipes.utils.dataset_utils import get_dataloader\n",
+    "from llama_cookbook.configs.datasets import samsum_dataset\n",
+    "from llama_cookbook.utils.dataset_utils import get_dataloader\n",
     "\n",
     "samsum_dataset.trust_remote_code = True\n",
     "\n",
@@ -248,7 +248,7 @@
    "source": [
     "from peft import get_peft_model, prepare_model_for_kbit_training, LoraConfig\n",
     "from dataclasses import asdict\n",
-    "from llama_recipes.configs import lora_config as LORA_CONFIG\n",
+    "from llama_cookbook.configs import lora_config as LORA_CONFIG\n",
     "\n",
     "lora_config = LORA_CONFIG()\n",
     "lora_config.r = 8\n",
@@ -278,7 +278,7 @@
    "outputs": [],
    "source": [
     "import torch.optim as optim\n",
-    "from llama_recipes.utils.train_utils import train\n",
+    "from llama_cookbook.utils.train_utils import train\n",
     "from torch.optim.lr_scheduler import StepLR\n",
     "\n",
     "model.train()\n",

getting-started/inference/local_inference/inference.py

@@ -10,9 +10,9 @@ import fire
 import torch
 
 from accelerate.utils import is_xpu_available
-from llama_recipes.inference.model_utils import load_model, load_peft_model
+from llama_cookbook.inference.model_utils import load_model, load_peft_model
 
-from llama_recipes.inference.safety_utils import AgentType, get_safety_checker
+from llama_cookbook.inference.safety_utils import AgentType, get_safety_checker
 from transformers import AutoTokenizer
 
 
@@ -176,7 +176,7 @@ def main(
                 )
             ],
             title="Meta Llama3 Playground",
-            description="https://github.com/meta-llama/llama-recipes",
+            description="https://github.com/meta-llama/llama-cookbook",
         ).queue().launch(server_name="0.0.0.0", share=share_gradio)
 
 

getting-started/inference/mobile_inference/android_inference/README.md

@@ -103,7 +103,7 @@ Connect your phone to your development machine. On OSX, you'll be prompted on th
 
 ## Building the Android Package with MLC
 
-First edit the file under `android/MLCChat/mlc-package-config.json` and with the [mlc-package-config.json](./mlc-package-config.json) in llama-recipes.
+First edit the file under `android/MLCChat/mlc-package-config.json` and with the [mlc-package-config.json](./mlc-package-config.json) in llama-cookbook.
 
 To understand what these JSON fields mean you can refer to this [documentation](https://llm.mlc.ai/docs/deploy/android.html#step-2-build-runtime-and-model-libraries).
 

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "llama-cookbook"
-version = "0.0.5"
+version = "0.0.5.post1"
 authors = [
   { name="Hamid Shojanazeri", email="hamidnazeri@meta.com" },
   { name="Matthias Reso", email="mreso@meta.com" },

src/llama_cookbook/configs/datasets.py

@@ -14,8 +14,8 @@ class samsum_dataset:
 @dataclass
 class grammar_dataset:
     dataset: str = "grammar_dataset"
-    train_split: str = "src/llama_recipes/datasets/grammar_dataset/gtrain_10k.csv"
-    test_split: str = "src/llama_recipes/datasets/grammar_dataset/grammar_validation.csv"
+    train_split: str = "src/llama_cookbook/datasets/grammar_dataset/gtrain_10k.csv"
+    test_split: str = "src/llama_cookbook/datasets/grammar_dataset/grammar_validation.csv"
 
 
 @dataclass
@@ -23,7 +23,7 @@ class alpaca_dataset:
     dataset: str = "alpaca_dataset"
     train_split: str = "train"
     test_split: str = "val"
-    data_path: str = "src/llama_recipes/datasets/alpaca_data.json"
+    data_path: str = "src/llama_cookbook/datasets/alpaca_data.json"
 
 @dataclass
 class custom_dataset:
@@ -32,7 +32,7 @@ class custom_dataset:
     train_split: str = "train"
     test_split: str = "validation"
     data_path: str = ""
-    
+
 @dataclass
 class llamaguard_toxicchat_dataset:
     dataset: str = "llamaguard_toxicchat_dataset"

src/llama_cookbook/configs/wandb.py

@@ -6,10 +6,10 @@ from dataclasses import dataclass, field
 
 @dataclass
 class wandb_config:
-    project: str = 'llama_recipes' # wandb project name
+    project: str = 'llama_cookbook' # wandb project name
     entity: Optional[str] = None # wandb entity name
     job_type: Optional[str] = None
     tags: Optional[List[str]] = None
     group: Optional[str] = None
     notes: Optional[str] = None
-    mode: Optional[str] = None
+    mode: Optional[str] = None

src/llama_cookbook/data/llama_guard/README.md

@@ -10,9 +10,9 @@ The finetuning_data_formatter script provides classes and methods for formatting
 
 ## Running the script
 
-1. Clone the llama-recipes repo
+1. Clone the llama-cookbook repo
 2. Install the dependencies
-3. Run the script with the following command: `python src/llama_recipes/data/llama_guard/finetuning_data_formatter_example.py > sample.json`
+3. Run the script with the following command: `python src/llama_cookbook/data/llama_guard/finetuning_data_formatter_example.py > sample.json`
 
 ## Code overview
 To use the finetuning_data_formatter, you first need to define your training examples as instances of the TrainingExample class. For example:

src/llama_cookbook/finetuning.py

@@ -74,7 +74,7 @@ def setup_wandb(train_config, fsdp_config, **kwargs):
             "You are trying to use wandb which is not currently installed. "
             "Please install it using pip install wandb"
         )
-    from llama_recipes.configs import wandb_config as WANDB_CONFIG
+    from llama_cookbook.configs import wandb_config as WANDB_CONFIG
 
     wandb_config = WANDB_CONFIG()
     update_config(wandb_config, **kwargs)
@@ -196,7 +196,7 @@ def main(**kwargs):
         model.resize_token_embeddings(len(tokenizer))
 
     print_model_size(model, train_config, rank if train_config.enable_fsdp else 0)
-    
+
     # Convert the model to bfloat16 if fsdp and pure_bf16 is enabled
     if (
         train_config.enable_fsdp
@@ -239,12 +239,12 @@ def main(**kwargs):
             freeze_transformer_layers(model, train_config.num_freeze_layers)
             # print model size and frozen layers after freezing layers
             print_frozen_model_status(model, train_config, rank if train_config.enable_fsdp else 0)
-            
+
         if not train_config.use_peft and train_config.freeze_LLM_only and config.model_type == "mllama":
             freeze_LLM_only(model)
             # print model size and frozen layers after freezing layers
             print_frozen_model_status(model, train_config, rank if train_config.enable_fsdp else 0)
-        
+
         mixed_precision_policy, wrapping_policy = get_policies(fsdp_config, rank)
         # Create the FSDP wrapper for MllamaSelfAttentionDecoderLayer,MllamaCrossAttentionDecoderLayer,MllamaVisionEncoderLayer in vision models
         if is_vision:
@@ -264,7 +264,7 @@ def main(**kwargs):
             device_id = torch.xpu.current_device()
         elif torch.cuda.is_available():
             device_id = torch.cuda.current_device()
-        
+
         if train_config.freeze_LLM_only:
             use_orig_params = True
         else:
@@ -312,7 +312,7 @@ def main(**kwargs):
         dataset_processer = processor
     else:
         dataset_processer = tokenizer
-    
+
     # Load and preprocess the dataset for training and validation
 
     dataset_train = get_preprocessed_dataset(

src/llama_cookbook/inference/safety_utils.py

@@ -24,7 +24,7 @@ class AuditNLGSensitiveTopics(object):
         except ImportError as e:
             print("Could not import optional dependency: auditnlg\nPlease install manually with:\n pip install auditnlg\nFollowed by:\npip install -r requirements.txt")
             raise e
-            
+
 
         data = [{"output": output_text}]
 
@@ -37,8 +37,8 @@ class AuditNLGSensitiveTopics(object):
             report += "|" + "|".join(f"{n:^10}" for n in [list(k.keys())[0] for k in scores["class_scores"]]) + "|\n"
             report += "|" + "|".join(f"{n:^10.5}" for n in [list(k.values())[0] for k in scores["class_scores"]]) + "|\n"
         return "Sensitive Topics", is_safe, report
-    
-    
+
+
 class SalesforceSafetyChecker(object):
     def __init__(self):
         pass
@@ -47,13 +47,13 @@ class SalesforceSafetyChecker(object):
         from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, AutoConfig
 
         config = AutoConfig.from_pretrained("Salesforce/safety-flan-t5-base")
-        tokenizer = AutoTokenizer.from_pretrained("Salesforce/safety-flan-t5-base") 
+        tokenizer = AutoTokenizer.from_pretrained("Salesforce/safety-flan-t5-base")
         model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/safety-flan-t5-base", config=config)
         model.eval()
 
         prefix = "Is the <Text> field safe or unsafe?"
         input_ids = tokenizer(prefix + " <Text> " + output_text + " <Context> ", return_tensors="pt").input_ids
-        
+
         if len(input_ids[0]) > 512:
             warnings.warn(
                 "Input length is > 512 token. Safety check result could be incorrect."
@@ -66,9 +66,9 @@ class SalesforceSafetyChecker(object):
                 return_dict_in_generate=True,
                 max_new_tokens=20,
                 )
-        
-        is_safe = tokenizer.decode(outputs.sequences[0], skip_special_tokens=True).split(" ")[0] == "safe"    
-            
+
+        is_safe = tokenizer.decode(outputs.sequences[0], skip_special_tokens=True).split(" ")[0] == "safe"
+
         report = ""
         if not is_safe:
             true_false_ids = tokenizer("true false").input_ids[:2]
@@ -76,11 +76,11 @@ class SalesforceSafetyChecker(object):
             scores = {}
             for k, i in zip(keys, range(3,20,2)):
                 scores[k] = round(outputs.scores[i][0,true_false_ids].softmax(dim=0)[0].item(), 5)
-            
+
             report += "|" + "|".join(f"{n:^10}" for n in scores.keys()) + "|\n"
             report += "|" + "|".join(f"{n:^10}" for n in scores.values()) + "|\n"
         return "Salesforce Content Safety Flan T5 Base", is_safe, report
-        
+
 
     def get_total_length(self, data):
         prefix = "Is the <Text> field safe or unsafe "
@@ -158,7 +158,7 @@ class LlamaGuardSafetyChecker(object):
 
     def __init__(self):
         from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
-        from llama_recipes.inference.prompt_format_utils import build_default_prompt, create_conversation, LlamaGuardVersion
+        from llama_cookbook.inference.prompt_format_utils import build_default_prompt, create_conversation, LlamaGuardVersion
 
         model_id = "meta-llama/Llama-Guard-3-8B"
 
@@ -168,7 +168,7 @@ class LlamaGuardSafetyChecker(object):
         self.model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config=quantization_config, device_map="auto")
 
     def __call__(self, output_text, **kwargs):
-        
+
         agent_type = kwargs.get('agent_type', AgentType.USER)
         user_prompt = kwargs.get('user_prompt', "")
 
@@ -194,14 +194,14 @@ class LlamaGuardSafetyChecker(object):
         prompt_len = input_ids.shape[-1]
         output = self.model.generate(input_ids=input_ids, max_new_tokens=100, pad_token_id=0)
         result = self.tokenizer.decode(output[0][prompt_len:], skip_special_tokens=True)
-        
+
         splitted_result = result.split("\n")[0];
-        is_safe = splitted_result == "safe"    
+        is_safe = splitted_result == "safe"
 
         report = result
-        
+
         return "Llama Guard", is_safe, report
-        
+
 
 # Function to load the PeftModel for performance optimization
 # Function to determine which safety checker to use based on the options selected

src/llama_cookbook/tools/README.md

@@ -7,7 +7,7 @@ This is the reverse conversion for `convert_llama_weights_to_hf.py` script from
 - Copy file params.json from the official llama download into that directory.
 - Run the conversion script. `model-path` can be a Hugging Face hub model or a local hf model directory.
 ```
-python -m llama_recipes.tools.convert_hf_weights_to_llama --model-path meta-llama/Meta-Llama-3.1-70B-Instruct --output-dir test70B --model-size 70B
+python -m llama_cookbook.tools.convert_hf_weights_to_llama --model-path meta-llama/Meta-Llama-3.1-70B-Instruct --output-dir test70B --model-size 70B
 ```
 
 ## Step 1: Run inference

src/llama_cookbook/utils/config_utils.py

@@ -49,10 +49,10 @@ def generate_peft_config(train_config, kwargs):
         raise RuntimeError(f"Peft config not found: {train_config.peft_method}")
 
     if train_config.peft_method == "prefix":
-        raise RuntimeError("PrefixTuning is currently not supported (see https://github.com/meta-llama/llama-recipes/issues/359#issuecomment-2089350811)")
+        raise RuntimeError("PrefixTuning is currently not supported (see https://github.com/meta-llama/llama-cookbook/issues/359#issuecomment-2089350811)")
 
     if train_config.enable_fsdp and train_config.peft_method == "llama_adapter":
-        raise RuntimeError("Llama_adapter is currently not supported in combination with FSDP (see https://github.com/meta-llama/llama-recipes/issues/359#issuecomment-2089274425)")
+        raise RuntimeError("Llama_adapter is currently not supported in combination with FSDP (see https://github.com/meta-llama/llama-cookbook/issues/359#issuecomment-2089274425)")
 
     config = configs[names.index(train_config.peft_method)]()