Upload 4 files

GPT2_TinyShakespeare1.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4034c8816050f8c3179c282399bdc3de05d9f4fb56014669df2b08a49a73f0d
+size 1544230233

app.py

@@ -0,0 +1,93 @@
+import tiktoken
+import os
+import torch
+from torch.nn import functional as F
+
+from model import GPTConfig, GPT
+import gradio as gr
+
+device = 'cpu'
+if torch.cuda.is_available():
+    device = 'cuda'
+elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available():
+    device = "mps"
+print(f"using device: {device}")
+
+modelpath = '.'
+
+# STOP
+max_length = 500
+
+enc = tiktoken.get_encoding('gpt2') 
+
+# CHANGES IN CURRENT CODE
+ckpt_path = os.path.join(modelpath, 'GPT2_TinyShakespeare1.pt')
+print(ckpt_path)
+checkpoint = torch.load(ckpt_path, map_location=device)
+gptconf = GPTConfig(**checkpoint['model_args'])
+model = GPT(gptconf)
+state_dict = checkpoint['model']
+unwanted_prefix = '_orig_mod.'
+for k,v in list(state_dict.items()):
+    if k.startswith(unwanted_prefix):
+        state_dict[k[len(unwanted_prefix):]] = state_dict.pop(k)
+model.load_state_dict(state_dict)
+
+model.to(device)
+model = torch.compile(model)
+
+def generateText(inputText, num_tokens=500):
+    start_tokens = enc.encode(inputText)
+    # print(start_tokens, len(start_tokens))
+    start_tokens = torch.tensor(start_tokens)
+    x = start_tokens.view(1, len(start_tokens))
+    x = x.to(device)
+    
+    while x.size(1) < max_length:
+        # forward the model to get the logits
+        with torch.no_grad():
+            logits = model(x)[0] # (B, T, vocab_size)
+            # take the logits at the last position
+            logits = logits[:, -1, :] # (B, vocab_size)
+            # get the probabilities
+            probs = F.softmax(logits, dim=-1)
+            # do top-k sampling of 50 (huggingface pipeline default)
+            # topk_probs here becomes (5, 50), topk_indices is (5, 50)
+            topk_probs, topk_indices = torch.topk(probs, 50, dim=-1)
+            # select a token from the top-k probabilities
+            # note: multinomial does not demand the input to sum to 1
+            ix = torch.multinomial(topk_probs, 1) # (B, 1)
+            # gather the corresponding indices
+            xcol = torch.gather(topk_indices, -1, ix) # (B, 1)
+            # append to the sequence
+            x = torch.cat((x, xcol), dim=1)
+            # print(x.size(1))
+    
+    tokens = x[0, :max_length].tolist()
+    decoded = enc.decode(tokens)
+    return decoded
+
+title = "GPT-2 Trained from Scratch"
+description = "GPT-2 trained on scratch on TinyShakespeare dataset"
+examples = [["ROMEO:\nWith love's light wings did I o'er-perch these walls;\nFor stony limits cannot hold love out,\nAnd what love can do that dares love attempt;\nTherefore thy kinsmen are no let to me.\n", 500],
+            ["CAPULET:\nWhy, how now, kinsman! wherefore storm you so?\n", 500],
+            ["KING RICHARD II:\nAy, hand from hand, my love, and heart from heart.\nAnd", 500],
+            ["QUEEN:\nBanish us both and send the king with me.\nAnd", 500],
+            ["CORIOLANUS:\n", 500]
+           ]
+
+demo = gr.Interface(
+    generateText, 
+    inputs = [
+        gr.Textbox(label="Starting text"),
+        gr.Slider(100, 2000, value = 500, step=100, label="Number of chars that you want in your output"),
+        ], 
+    outputs = [
+        gr.Text(),
+        ],
+    title = title,
+    description = description,
+    examples = examples,
+    cache_examples=False    
+)
+demo.launch()

model.py

@@ -0,0 +1,159 @@
+from dataclasses import dataclass
+import inspect
+import torch
+import torch.nn as nn
+from torch.nn import functional as F
+
+class CausalSelfAttention(nn.Module):
+    """ multiple heads of self-attention in parallel """
+
+    def __init__(self, config):
+        super().__init__()
+        assert config.n_embd % config.n_head ==0
+        self.c_attn = nn.Linear(config.n_embd, 3 * config.n_embd)
+        self.c_proj = nn.Linear(config.n_embd, config.n_embd)
+        self.c_proj.NANOGPT_SCALE_INIT = 1
+        self.n_head = config.n_head
+        self.n_embd = config.n_embd
+        self.register_buffer("bias", torch.tril(torch.ones(config.block_size, config.block_size))
+                             .view(1, 1, config.block_size, config.block_size))
+
+    def forward(self, x):
+        B, T, C = x.size() # batch size, sequence length, embedding dimensionality (n_embd)
+        # calculate query, key, values for all heads in batch and move head forward to be the batch dim
+        # nh is "number of heads", hs is "head size", and C (number of channels) = nh * hs
+        # e.g. in GPT-2 (124M), n_head=12, hs=64, so nh*hs=C=768 channels in the Transformer
+        qkv = self.c_attn(x)
+        q, k, v = qkv.split(self.n_embd, dim=2)
+        k = k.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
+        q = q.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
+        v = v.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
+
+        # att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))
+        # att = att.masked_fill(self.bias[:, :, :T, :T] == 0, float('-inf'))
+        # att = F.softmax(att, dim=-1)
+        # y = att @ v # (B, nh, T, T) x (B, nh, T, hs) -> (B, nh, T, hs)
+
+        y = F.scaled_dot_product_attention(q, k, v, is_causal = True) # Flash attention
+
+        y = y.transpose(1, 2).contiguous().view(B, T, C) # re-assemble all head outputs side by side
+        # output projection
+        y = self.c_proj(y)
+        return y
+
+
+class MLP(nn.Module):
+    """ a simple linear layer followed by a non-linearity """
+
+    def __init__(self, config):
+        super().__init__()
+        self.c_fc = nn.Linear(config.n_embd, 4*config.n_embd)
+        self.gelu = nn.GELU(approximate='tanh')
+        self.c_proj = nn.Linear(4*config.n_embd, config.n_embd)
+        self.c_proj.NANOGPT_SCALE_INIT = 1
+
+    def forward(self, x):
+        x = self.c_fc(x)
+        x = self.gelu(x)
+        x = self.c_proj(x)
+        return x
+
+
+class Block(nn.Module):
+
+    def __init__(self, config):
+        super().__init__()
+        self.ln_1 = nn.LayerNorm(config.n_embd)
+        self.attn = CausalSelfAttention(config)
+        self.ln_2 = nn.LayerNorm(config.n_embd)
+        self.mlp = MLP(config)
+
+    def forward(self, x):
+        x = x + self.attn(self.ln_1(x))
+        x = x + self.mlp(self.ln_2(x))
+        return x
+
+@dataclass
+class GPTConfig:
+    block_size: int = 1024 # max sequence length
+    vocab_size: int = 50304 # number of tokens: 50,000 BPE merges + 256 bytes tokens + 1 <|endoftext|> token
+    n_layer: int = 12 # number of layers
+    n_head: int = 12 # number of heads
+    n_embd: int = 768 # embedding dimension
+
+class GPT(nn.Module):
+
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+
+        self.transformer = nn.ModuleDict(dict(
+            wte = nn.Embedding(config.vocab_size, config.n_embd),
+            wpe = nn.Embedding(config.block_size, config.n_embd),
+            h = nn.ModuleList([Block(config) for _ in range(config.n_layer)]),
+            ln_f = nn.LayerNorm(config.n_embd),
+        ))
+        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
+
+        # weight sharing
+        self.transformer.wte.weight = self.lm_head.weight
+
+        # weight initialization
+        self.apply(self._init_weights)
+
+    def _init_weights(self, module):
+        if isinstance(module, nn.Linear):
+            std = 0.02
+            if hasattr(module, 'NANOGPT_SCALE_INIT'):
+                std *= (2 * self.config.n_layer) ** -0.5
+            torch.nn.init.normal_(module.weight, mean = 0.0, std = std)
+            if module.bias is not None:
+                torch.nn.init.zeros_(module.bias)
+        elif isinstance(module, nn.Embedding):
+            torch.nn.init.normal_(module.weight, mean=0.0, std = 0.02)
+
+
+    def forward(self, idx, targets=None):
+        # idx is of shape (B, T)
+        B, T = idx.size()
+        assert T <= self.config.block_size, f"Cannot forward sequence of length {T}, block size is only {self.config.block_size}"
+        # forward the token and posisition embeddings
+        pos = torch.arange(0, T, dtype=torch.long, device=idx.device) # shape (T)
+        pos_emb = self.transformer.wpe(pos) # position embeddings of shape (T, n_embd)
+        tok_emb = self.transformer.wte(idx) # token embeddings of shape (B, T, n_embd)
+        x = tok_emb + pos_emb
+        # forward the blocks of the transformer
+        for block in self.transformer.h:
+            x = block(x)
+        # forward the final layernorm and the classifier
+        x = self.transformer.ln_f(x)
+        logits = self.lm_head(x) # (B, T, vocab_size)
+        loss = None
+        if targets is not None:
+            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
+        return logits, loss
+
+    def configure_optimizers(self, weight_decay, learning_rate, device_type):
+        # start with all of the candidate parameters (that require grad)
+        param_dict = {pn: p for pn, p in self.named_parameters()}
+        param_dict = {pn: p for pn, p in param_dict.items() if p.requires_grad}
+        # create optim groups. Any parameters that is 2D will be weight decayed, otherwise no.
+        # i.e. all weight tensors in matmuls + embeddings decay, all biases and layernorms don't.
+        decay_params = [p for n, p in param_dict.items() if p.dim() >= 2]
+        nodecay_params = [p for n, p in param_dict.items() if p.dim() < 2]
+        optim_groups = [
+            {'params': decay_params, 'weight_decay': weight_decay},
+            {'params': nodecay_params, 'weight_decay': 0.0}
+        ]
+        num_decay_params = sum(p.numel() for p in decay_params)
+        num_nodecay_params = sum(p.numel() for p in nodecay_params)
+        
+        print(f"num decayed parameter tensors: {len(decay_params)}, with {num_decay_params:,} parameters")
+        print(f"num non-decayed parameter tensors: {len(nodecay_params)}, with {num_nodecay_params:,} parameters")
+        # Create AdamW optimizer and use the fused version if it is available
+        fused_available = 'fused' in inspect.signature(torch.optim.AdamW).parameters
+        use_fused = fused_available and device_type == "cuda"
+        
+        print(f"using fused AdamW: {use_fused}")
+        optimizer = torch.optim.AdamW(optim_groups, lr=learning_rate, betas=(0.9, 0.95), eps=1e-8, fused=use_fused)
+        return optimizer

requirements.txt

@@ -0,0 +1,2 @@
+torch
+tiktoken