flops analysis

ch04/01_main-chapter-code/README.md

@@ -1,4 +1,4 @@
-# Chapter 4: Implementing a GPT model from Scratch To Generate Text
+# Chapter 4: Implementing a GPT Model from Scratch To Generate Text
 
 - [ch04.ipynb](ch04.ipynb) contains all the code as it appears in the chapter
 - [previous_chapters.py](previous_chapters.py) is a Python module that contains the `MultiHeadAttention` module from the previous chapter, which we import in [ch04.ipynb](ch04.ipynb) to create the GPT model

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

@@ -1489,7 +1489,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.6"
+   "version": "3.11.4"
   }
  },
  "nbformat": 4,

ch04/02_performance-analysis/README.md

@@ -0,0 +1,5 @@
+# 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

@@ -0,0 +1,154 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<font size=\"1\">\n",
+    "Supplementary code for \"Build a Large Language Model From Scratch\": <a href=\"https://www.manning.com/books/build-a-large-language-model-from-scratch\">https://www.manning.com/books/build-a-large-language-model-from-scratch</a> by <a href=\"https://sebastianraschka.com\">Sebastian Raschka</a><br>\n",
+    "Code repository: <a href=\"https://github.com/rasbt/LLMs-from-scratch\">https://github.com/rasbt/LLMs-from-scratch</a>\n",
+    "</font>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## FLOPS Analysis"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- FLOPs (Floating Point Operations Per Second) measure the computational complexity of neural network models by counting the number of floating-point operations executed\n",
+    "-  High FLOPs indicate more intensive computation and energy consumption"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# pip install -r requirements-extra.txt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "thop version: 0.1.1-2209072238\n",
+      "torch version: 2.2.2\n",
+      "tiktoken version: 0.5.1\n"
+     ]
+    }
+   ],
+   "source": [
+    "from importlib.metadata import version\n",
+    "\n",
+    "import matplotlib\n",
+    "import tiktoken\n",
+    "import torch\n",
+    "\n",
+    "print(\"thop version:\", version(\"thop\"))\n",
+    "print(\"torch version:\", version(\"torch\"))\n",
+    "print(\"tiktoken version:\", version(\"tiktoken\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "GerIdRMXd6g9",
+    "outputId": "ccdd5c71-d221-4a84-f9bc-09557e77162d"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "gpt-small (124M)  : 5.1e+11 FLOPS\n",
+      "gpt-medium (355M) : 1.4e+12 FLOPS\n",
+      "gpt-large (774M)  : 3.2e+12 FLOPS\n",
+      "gpt-xl (1558M)    : 6.4e+12 FLOPS\n"
+     ]
+    }
+   ],
+   "source": [
+    "import torch\n",
+    "from thop import profile\n",
+    "\n",
+    "from previous_chapters import GPTModel\n",
+    "\n",
+    "\n",
+    "BASE_CONFIG = {\n",
+    "    \"vocab_size\": 50257,     # Vocabulary size\n",
+    "    \"context_length\": 1024,  # Context length\n",
+    "    \"drop_rate\": 0.0,        # Dropout rate\n",
+    "    \"qkv_bias\": True         # Query-key-value bias\n",
+    "}\n",
+    "\n",
+    "model_configs = {\n",
+    "    \"gpt-small (124M)\": {\"emb_dim\": 768, \"n_layers\": 12, \"n_heads\": 12},\n",
+    "    \"gpt-medium (355M)\": {\"emb_dim\": 1024, \"n_layers\": 24, \"n_heads\": 16},\n",
+    "    \"gpt-large (774M)\": {\"emb_dim\": 1280, \"n_layers\": 36, \"n_heads\": 20},\n",
+    "    \"gpt-xl (1558M)\": {\"emb_dim\": 1600, \"n_layers\": 48, \"n_heads\": 25},\n",
+    "}\n",
+    "\n",
+    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
+    "input_tensor = torch.randint(0, 50257, (2, 1024)).to(device)\n",
+    "\n",
+    "for size in model_configs:\n",
+    "    BASE_CONFIG.update(model_configs[size])\n",
+    "    \n",
+    "    model = GPTModel(BASE_CONFIG).bfloat16()\n",
+    "    model.to(device)\n",
+    "\n",
+    "    # MACS = multiply-accumulate operations\n",
+    "    # MACS are typically counted as two FLOPS (one multiply and one accumulate)\n",
+    "    macs, params = profile(model, inputs=(input_tensor,), verbose=False)\n",
+    "    flops = 2*macs\n",
+    "    print(f\"{size:18}: {flops:.1e} FLOPS\")\n",
+    "    \n",
+    "    del model\n",
+    "    torch.cuda.empty_cache()"
+   ]
+  }
+ ],
+ "metadata": {
+  "accelerator": "GPU",
+  "colab": {
+   "gpuType": "A100",
+   "machine_shape": "hm",
+   "provenance": []
+  },
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

ch04/02_performance-analysis/previous_chapters.py

@@ -0,0 +1,279 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+#
+# This file collects all the relevant code that we covered thus far
+# throughout Chapters 2-4.
+# This file can be run as a standalone script.
+
+import tiktoken
+import torch
+import torch.nn as nn
+from torch.utils.data import Dataset, DataLoader
+
+#####################################
+# Chapter 2
+#####################################
+
+
+class GPTDatasetV1(Dataset):
+    def __init__(self, txt, tokenizer, max_length, stride):
+        self.input_ids = []
+        self.target_ids = []
+
+        # Tokenize the entire text
+        token_ids = tokenizer.encode(txt)
+
+        # Use a sliding window to chunk the book into overlapping sequences of max_length
+        for i in range(0, len(token_ids) - max_length, stride):
+            input_chunk = token_ids[i:i + max_length]
+            target_chunk = token_ids[i + 1: i + max_length + 1]
+            self.input_ids.append(torch.tensor(input_chunk))
+            self.target_ids.append(torch.tensor(target_chunk))
+
+    def __len__(self):
+        return len(self.input_ids)
+
+    def __getitem__(self, idx):
+        return self.input_ids[idx], self.target_ids[idx]
+
+
+def create_dataloader_v1(txt, batch_size=4, max_length=256,
+                         stride=128, shuffle=True, drop_last=True, num_workers=0):
+    # Initialize the tokenizer
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    # Create dataset
+    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
+
+    # Create dataloader
+    dataloader = DataLoader(
+        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=0)
+
+    return dataloader
+
+
+#####################################
+# Chapter 3
+#####################################
+class MultiHeadAttention(nn.Module):
+    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
+        super().__init__()
+        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
+
+        self.d_out = d_out
+        self.num_heads = num_heads
+        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
+
+        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
+        self.dropout = nn.Dropout(dropout)
+        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
+
+    def forward(self, x):
+        b, num_tokens, d_in = x.shape
+
+        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
+        queries = self.W_query(x)
+        values = self.W_value(x)
+
+        # We implicitly split the matrix by adding a `num_heads` dimension
+        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
+        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
+        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
+        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
+
+        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
+        keys = keys.transpose(1, 2)
+        queries = queries.transpose(1, 2)
+        values = values.transpose(1, 2)
+
+        # Compute scaled dot-product attention (aka self-attention) with a causal mask
+        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
+
+        # Original mask truncated to the number of tokens and converted to boolean
+        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
+
+        # Use the mask to fill attention scores
+        attn_scores.masked_fill_(mask_bool, -torch.inf)
+
+        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
+        attn_weights = self.dropout(attn_weights)
+
+        # Shape: (b, num_tokens, num_heads, head_dim)
+        context_vec = (attn_weights @ values).transpose(1, 2)
+
+        # Combine heads, where self.d_out = self.num_heads * self.head_dim
+        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
+        context_vec = self.out_proj(context_vec)  # optional projection
+
+        return context_vec
+
+
+#####################################
+# Chapter 4
+#####################################
+class LayerNorm(nn.Module):
+    def __init__(self, emb_dim):
+        super().__init__()
+        self.eps = 1e-5
+        self.scale = nn.Parameter(torch.ones(emb_dim))
+        self.shift = nn.Parameter(torch.zeros(emb_dim))
+
+    def forward(self, x):
+        mean = x.mean(dim=-1, keepdim=True)
+        var = x.var(dim=-1, keepdim=True, unbiased=False)
+        norm_x = (x - mean) / torch.sqrt(var + self.eps)
+        return self.scale * norm_x + self.shift
+
+
+class GELU(nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, x):
+        return 0.5 * x * (1 + torch.tanh(
+            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
+            (x + 0.044715 * torch.pow(x, 3))
+        ))
+
+
+class FeedForward(nn.Module):
+    def __init__(self, cfg):
+        super().__init__()
+        self.layers = nn.Sequential(
+            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
+            GELU(),
+            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
+        )
+
+    def forward(self, x):
+        return self.layers(x)
+
+
+class TransformerBlock(nn.Module):
+    def __init__(self, cfg):
+        super().__init__()
+        self.att = MultiHeadAttention(
+            d_in=cfg["emb_dim"],
+            d_out=cfg["emb_dim"],
+            context_length=cfg["context_length"],
+            num_heads=cfg["n_heads"],
+            dropout=cfg["drop_rate"],
+            qkv_bias=cfg["qkv_bias"])
+        self.ff = FeedForward(cfg)
+        self.norm1 = LayerNorm(cfg["emb_dim"])
+        self.norm2 = LayerNorm(cfg["emb_dim"])
+        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
+
+    def forward(self, x):
+        # Shortcut connection for attention block
+        shortcut = x
+        x = self.norm1(x)
+        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
+        x = self.drop_shortcut(x)
+        x = x + shortcut  # Add the original input back
+
+        # Shortcut connection for feed-forward block
+        shortcut = x
+        x = self.norm2(x)
+        x = self.ff(x)
+        x = self.drop_shortcut(x)
+        x = x + shortcut  # Add the original input back
+
+        return x
+
+
+class GPTModel(nn.Module):
+    def __init__(self, cfg):
+        super().__init__()
+        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
+        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
+        self.drop_emb = nn.Dropout(cfg["drop_rate"])
+
+        self.trf_blocks = nn.Sequential(
+            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
+
+        self.final_norm = LayerNorm(cfg["emb_dim"])
+        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
+
+    def forward(self, in_idx):
+        batch_size, seq_len = in_idx.shape
+        tok_embeds = self.tok_emb(in_idx)
+        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
+        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
+        x = self.drop_emb(x)
+        x = self.trf_blocks(x)
+        x = self.final_norm(x)
+        logits = self.out_head(x)
+        return logits
+
+
+def generate_text_simple(model, idx, max_new_tokens, context_size):
+    # idx is (B, T) array of indices in the current context
+    for _ in range(max_new_tokens):
+
+        # Crop current context if it exceeds the supported context size
+        # E.g., if LLM supports only 5 tokens, and the context size is 10
+        # then only the last 5 tokens are used as context
+        idx_cond = idx[:, -context_size:]
+
+        # Get the predictions
+        with torch.no_grad():
+            logits = model(idx_cond)
+
+        # Focus only on the last time step
+        # (batch, n_token, vocab_size) becomes (batch, vocab_size)
+        logits = logits[:, -1, :]
+
+        # Get the idx of the vocab entry with the highest logits value
+        idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch, 1)
+
+        # Append sampled index to the running sequence
+        idx = torch.cat((idx, idx_next), dim=1)  # (batch, n_tokens+1)
+
+    return idx
+
+
+if __name__ == "__main__":
+
+    GPT_CONFIG_124M = {
+        "vocab_size": 50257,     # Vocabulary size
+        "context_length": 1024,  # Context length
+        "emb_dim": 768,          # Embedding dimension
+        "n_heads": 12,           # Number of attention heads
+        "n_layers": 12,          # Number of layers
+        "drop_rate": 0.1,        # Dropout rate
+        "qkv_bias": False        # Query-Key-Value bias
+    }
+
+    torch.manual_seed(123)
+    model = GPTModel(GPT_CONFIG_124M)
+    model.eval()  # disable dropout
+
+    start_context = "Hello, I am"
+
+    tokenizer = tiktoken.get_encoding("gpt2")
+    encoded = tokenizer.encode(start_context)
+    encoded_tensor = torch.tensor(encoded).unsqueeze(0)
+
+    print(f"\n{50*'='}\n{22*' '}IN\n{50*'='}")
+    print("\nInput text:", start_context)
+    print("Encoded input text:", encoded)
+    print("encoded_tensor.shape:", encoded_tensor.shape)
+
+    out = generate_text_simple(
+        model=model,
+        idx=encoded_tensor,
+        max_new_tokens=10,
+        context_size=GPT_CONFIG_124M["context_length"]
+    )
+    decoded_text = tokenizer.decode(out.squeeze(0).tolist())
+
+    print(f"\n\n{50*'='}\n{22*' '}OUT\n{50*'='}")
+    print("\nOutput:", out)
+    print("Output length:", len(out[0]))
+    print("Output text:", decoded_text)

ch04/02_performance-analysis/requirements-extra.txt

@@ -0,0 +1 @@
+thop

ch04/README.md

@@ -1,3 +1,4 @@
 # Chapter 4: Implementing a GPT Model from Scratch to Generate Text
 
-- [01_main-chapter-code](01_main-chapter-code) contains the main chapter code.
+- [01_main-chapter-code](01_main-chapter-code) contains the main chapter code.
+- [02_performance-analysis](02_performance-analysis) contains optional code analyzing the performance of the GPT model(s) implemented in the main chapter.