Add readme (#577)

Author: rasbt

ch05/README.md

@@ -16,4 +16,11 @@
 - [07_gpt_to_llama](07_gpt_to_llama) contains a step-by-step guide for converting a GPT architecture implementation to Llama 3.2 and loads pretrained weights from Meta AI
 - [08_memory_efficient_weight_loading](08_memory_efficient_weight_loading) contains a bonus notebook showing how to load model weights via PyTorch's `load_state_dict` method more efficiently
 - [09_extending-tokenizers](09_extending-tokenizers) contains a from-scratch implementation of the GPT-2 BPE tokenizer
-- [10_llm-training-speed](10_llm-training-speed) shows PyTorch performance tips to improve the LLM training speed
+- [10_llm-training-speed](10_llm-training-speed) shows PyTorch performance tips to improve the LLM training speed
+
+
+
+<br>
+<br>
+
+[![Link to the video](https://img.youtube.com/vi/Zar2TJv-sE0/0.jpg)](https://www.youtube.com/watch?v=Zar2TJv-sE0)

pkg/llms_from_scratch/README.md

@@ -0,0 +1,110 @@
+# `llms-from-scratch` PyPI Package
+
+This optional PyPI package lets you conveniently import code from various chapters of the *Build a Large Language Model From Scratch* book.
+
+ 
+## Installation
+
+ 
+### From PyPI
+
+Install the `llms-from-scratch` package from the official [Python Package Index](https://pypi.org/project/llms-from-scratch/) (PyPI):
+
+```bash
+pip install llms-from-scratch
+```
+
+> **Note:** If you're using [`uv`](https://github.com/astral-sh/uv), replace `pip` with `uv pip` or use `uv add`:
+
+```bash
+uv add llms-from-scratch
+```
+
+
+
+ 
+### Editable Install from GitHub
+
+If you'd like to modify the code and have those changes reflected during development:
+
+```bash
+git clone https://github.com/rasbt/LLMs-from-scratch.git
+cd LLMs-from-scratch
+pip install -e .
+```
+
+> **Note:** With `uv`, use:
+
+```bash
+uv add --editable . --dev
+```
+
+
+
+ 
+## Using the Package
+
+Once installed, you can import code from any chapter using:
+
+```python
+from llms_from_scratch.ch02 import GPTDatasetV1, create_dataloader_v1
+
+from llms_from_scratch.ch03 import (
+    MultiHeadAttention,
+    SelfAttention_v1,
+    SelfAttention_v2,
+    CausalAttention,
+    MultiHeadAttentionWrapper
+)
+
+from llms_from_scratch.ch04 import (
+    LayerNorm,
+    GELU,
+    FeedForward,
+    TransformerBlock,
+    GPTModel,
+    generate_text_simple
+)
+
+from llms_from_scratch.ch05 import (
+    generate,
+    train_model_simple,
+    evaluate_model,
+    generate_and_print_sample,
+    assign,
+    load_weights_into_gpt,
+    text_to_token_ids,
+    token_ids_to_text,
+    calc_loss_batch,
+    calc_loss_loader,
+    plot_losses
+)
+
+from llms_from_scratch.ch06 import (
+    download_and_unzip_spam_data,
+    create_balanced_dataset,
+    random_split,
+    SpamDataset,
+    calc_accuracy_loader,
+    evaluate_model,
+    train_classifier_simple,
+    plot_values,
+    classify_review
+)
+
+from llms_from_scratch.ch07 import (
+    download_and_load_file,
+    format_input,
+    InstructionDataset,
+    custom_collate_fn,
+    check_if_running,
+    query_model,
+    generate_model_scores
+)
+
+	
+from llms_from_scratch.appendix_a import NeuralNetwork, ToyDataset
+
+from llms_from_scratch.appendix_d import find_highest_gradient, train_model
+```
+