Add PyPI package (#576)

* Add PyPI package

* fixes

* fixes

Author: rasbt

.github/workflows/basic-tests-linux-uv.yml

@@ -66,3 +66,9 @@ jobs:
         run: |
           source .venv/bin/activate
           pytest ch02/05_bpe-from-scratch/tests/tests.py
+
+      - name: Test Selected Bonus Materials
+        shell: bash
+        run: |
+          source .venv/bin/activate
+          pytest pkg/llms_from_scratch/tests/

appendix-D/01_main-chapter-code/appendix-D.ipynb

@@ -68,6 +68,11 @@
     "\n",
     "\n",
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "GPT_CONFIG_124M = {\n",
     "    \"vocab_size\": 50257,   # Vocabulary size\n",
@@ -139,6 +144,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import create_dataloader_v1\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch02 import create_dataloader_v1\n",
+    "\n",
     "\n",
     "# Train/validation ratio\n",
     "train_ratio = 0.90\n",
@@ -454,6 +462,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import calc_loss_batch\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import calc_loss_batch\n",
+    "\n",
     "\n",
     "torch.manual_seed(123)\n",
     "model = GPTModel(GPT_CONFIG_124M)\n",
@@ -551,6 +562,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import evaluate_model, generate_and_print_sample\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import evaluate_model, generate_and_print_samplee\n",
+    "\n",
     "\n",
     "ORIG_BOOK_VERSION = False\n",
     "\n",
@@ -790,6 +804,9 @@
    ],
    "source": [
     "from previous_chapters import plot_losses\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import plot_losses\n",
+    "\n",
     "\n",
     "epochs_tensor = torch.linspace(1, n_epochs, len(train_losses))\n",
     "plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses)\n",
@@ -823,7 +840,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

appendix-E/01_main-chapter-code/appendix-E.ipynb

@@ -198,6 +198,16 @@
     "    create_balanced_dataset,\n",
     "    random_split\n",
     ")\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch06 import (\n",
+    "#     download_and_unzip_spam_data,\n",
+    "#     create_balanced_dataset,\n",
+    "#     random_split\n",
+    "# )\n",
+    "\n",
     "\n",
     "\n",
     "url = \"https://archive.ics.uci.edu/static/public/228/sms+spam+collection.zip\"\n",
@@ -409,6 +419,10 @@
    "source": [
     "from gpt_download import download_and_load_gpt2\n",
     "from previous_chapters import GPTModel, load_weights_into_gpt\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
+    "# from llms_from_scratch.ch05 import load_weights_into_gpt\n",
+    "\n",
     "\n",
     "\n",
     "CHOOSE_MODEL = \"gpt2-small (124M)\"\n",
@@ -577,6 +591,9 @@
    ],
    "source": [
     "from previous_chapters import calc_accuracy_loader\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import calc_accuracy_loader\n",
+    "\n",
     "\n",
     "\n",
     "torch.manual_seed(123)\n",
@@ -1387,6 +1404,8 @@
    "source": [
     "import time\n",
     "from previous_chapters import train_classifier_simple\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import train_classifier_simple\n",
     "\n",
     "\n",
     "start_time = time.time()\n",
@@ -1442,6 +1461,8 @@
    ],
    "source": [
     "from previous_chapters import plot_values\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import plot_values\n",
     "\n",
     "epochs_tensor = torch.linspace(0, num_epochs, len(train_losses))\n",
     "examples_seen_tensor = torch.linspace(0, examples_seen, len(train_losses))\n",

ch04/01_main-chapter-code/ch04.ipynb

@@ -970,6 +970,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch03 import MultiHeadAttention\n",
+    "\n",
     "from previous_chapters import MultiHeadAttention\n",
     "\n",
     "\n",

ch04/02_performance-analysis/README.md

@@ -1,5 +1,4 @@
 # Chapter 4: Implementing a GPT Model from Scratch To Generate Text
 
 - [flops-analysis.ipynb](flops-analysis.ipynb) analyses the floating point operations per second (FLOPS) of the GPT model(s) implemented in the main chapter. 
-- [previous_chapters.py](previous_chapters.py) is a Python module containing the `GPTModel` code we implemented in chapter 4 and other code implemented in previous chapters, which we import in the analysis notebook.
 - `requirements-extra.txt` includes additional Python libraries that need to be installed (via `pip install -r requirements-extra.txt`.

ch04/02_performance-analysis/flops-analysis.ipynb

@@ -127,7 +127,9 @@
     "import torch\n",
     "from thop import profile\n",
     "\n",
-    "from previous_chapters import GPTModel\n",
+    "# For installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "\n",
     "BASE_CONFIG = {\n",
@@ -550,7 +552,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch05/01_main-chapter-code/ch05.ipynb

@@ -147,6 +147,11 @@
    "source": [
     "import torch\n",
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "GPT_CONFIG_124M = {\n",
     "    \"vocab_size\": 50257,   # Vocabulary size\n",
@@ -212,6 +217,9 @@
     "import tiktoken\n",
     "from previous_chapters import generate_text_simple\n",
     "\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch04 import generate_text_simple\n",
+    "\n",
     "def text_to_token_ids(text, tokenizer):\n",
     "    encoded = tokenizer.encode(text, allowed_special={'<|endoftext|>'})\n",
     "    encoded_tensor = torch.tensor(encoded).unsqueeze(0) # add batch dimension\n",
@@ -924,6 +932,8 @@
    "outputs": [],
    "source": [
     "from previous_chapters import create_dataloader_v1\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch02 import create_dataloader_v1\n",
     "\n",
     "# Train/validation ratio\n",
     "train_ratio = 0.90\n",
@@ -2548,7 +2558,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.8"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,