Upload folder using huggingface_hub

model_gpt2.py

@@ -0,0 +1,242 @@
+from dataclasses import dataclass
+import torch
+import torch.nn as nn
+from torch.nn import functional as F
+import inspect
+
+#------------------------------------------
+
+class CausalSelfAttention(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        # key, query, value projection for al heads but in a batch
+        self.c_attn = nn.Linear(config.n_embd, 3 * config.n_embd)
+        # output projection
+        self.c_proj = nn.Linear(config.n_embd, config.n_embd)
+        self.c_proj.NANOGPT_SCALE_INIT = True
+
+        # regularization
+        self.n_head = config.n_head
+        self.n_embd = config.n_embd
+
+        # not really a 'bias', more of a mask, but following a openAI/HF naming
+        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_embed)
+        # 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)
+
+        # attention (materilizes the large (T,T) matrix for all the queries and keys)
+        # att = (q @ k.transpose(-2, -1)) * (1.0 / torch.sqrt(torch.tensor(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) @ (B, nh, T, hs) -> (B, nh, T, T)
+        # 4 lines above replaced by flash- attention
+        y = F.scaled_dot_product_attention(q, k, v, is_causal=True)
+
+        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):
+    def __init__(self, config):
+        super().__init__()
+        self.c_fc = nn.Linear(config.n_embd, 4 * config.n_embd)
+        self.gelu = nn.GELU(approximate='tanh') # historic reason for approximation
+        self.c_proj = nn.Linear(4 * config.n_embd, config.n_embd)
+        self.c_proj.NANOGPT_SCALE_INIT = True
+
+    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 lenghts
+    vocab_size: int = 50257 # number of tokens, 50,000 BPE merges + 256 byte tokens + 1 <|endoftext|>
+    n_layer: int = 12 # number of layers
+    n_head: int = 12 # number of heads
+    n_embd: int = 768 # embedding dim
+
+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 scheme
+        self.transformer.wte.weight = self.lm_head.weight
+
+        # init
+        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 # 2 times as each layer has attention and MLP
+            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) # word embedding will be initialised twice
+
+    def forward(self, idx, target = None):
+        # idx 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 thetoken and position embedding
+        pos = torch.arange(0, T, dtype = torch.long, device=idx.device) # (T)
+        pos_emb = self.transformer.wpe(pos) # (T, C)
+        tok_emb = self.transformer.wte(idx) # (B, T, C)
+        x = tok_emb + pos_emb # (B, T, C)
+        # forward the block for transformer
+        for block in self.transformer.h:
+            x = block(x)
+        # forward the final layer nor and classifier
+        x = self.transformer.ln_f(x)
+        logits = self.lm_head(x) # (B, T, vocab_size)
+        # compute the loss
+        loss = None
+        if target is not None:
+            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), target.view(-1))
+        return logits, loss
+
+    @classmethod
+    def from_pretrained(cls, model_type):
+        """ Loads pretrained GPT2 model from HuggingFace """
+        assert model_type in {"gpt2", "gpt2-medium", "gpt2-large", "gpt2-xl"}
+        from transformers import GPT2LMHeadModel
+        print(f"Loading {model_type} model...")
+
+        config_args = {
+            "gpt2": dict(n_layer = 12, n_head = 12, n_embd = 768), # 124M
+            "gpt2-medium": dict(n_layer = 24, n_head = 16, n_embd = 1024), # 350M
+            "gpt2-large": dict(n_layer = 36, n_head = 20, n_embd = 1280), # 774M
+            "gpt2-xl": dict(n_layer = 48, n_head = 25, n_embd = 1600), # 1558M
+        }[model_type]
+
+        config_args["vocab_size"] = 50257 # always for GPT2 checkpoints
+        config_args["block_size"] = 1024 # always for GPT2 checkpoints
+
+        config = GPTConfig(**config_args)
+        model = GPT(config)
+        sd = model.state_dict()
+        sd_keys = sd.keys()
+
+        sd_keys = [k for k in sd_keys if not k.endswith(".attn.bias")] # discard this mask
+
+        # init hugging face model
+        model_hf = GPT2LMHeadModel.from_pretrained(model_type)
+        sd_hf = model_hf.state_dict()
+        sd_keys_hf = sd_hf.keys()
+
+        # copy while ensuring all of the parameters are aligned and match in names and types
+        sd_keys_hf = [k for k in sd_keys_hf if not k.endswith(".attn.masked_bias")] # ignore these, just a buffer
+        sd_keys_hf = [k for k in sd_keys_hf if not k.endswith(".attn.bias")] # same, just the mask (buffer)
+        transposed = ['attn.c_attn.weight', 'attn.c_proj.weight', 'mlp.c_fc.weight', 'mlp.c_proj.weight']
+
+        # basically the openai checkpoints use a "Conv1D" module, but we only want to use a vanilla Linear
+        # this means that we have to transpose these weights when we import them
+        assert len(sd_keys_hf) == len(sd_keys), f"mismatched keys: {len(sd_keys_hf)} != {len(sd_keys)}"
+        for k in sd_keys_hf:
+            if any(k.endswith(w) for w in transposed):
+                # special treatment for the Conv1D weights we need to transpose
+                assert sd_hf[k].shape[::-1] == sd[k].shape
+                with torch.no_grad():
+                    sd[k].copy_(sd_hf[k].t())
+            else:
+                # vanilla copy over the other parameters
+                assert sd_hf[k].shape == sd[k].shape
+                with torch.no_grad():
+                    sd[k].copy_(sd_hf[k])
+
+        return model
+
+    def configure_optimizers(self, weight_decay, learning_rate, betas, device_type):
+        # start with all of the candidate parameters
+        param_dict = {pn: p for pn, p in self.named_parameters()}
+        # filter out those that do not require grad
+        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'
+        extra_args = dict(fused=True) if use_fused else dict()
+        optimizer = torch.optim.AdamW(optim_groups, lr=learning_rate, betas=betas, **extra_args)
+        print(f"using fused AdamW: {use_fused}")
+
+        return optimizer
+
+    @torch.no_grad()
+    def generate(self, idx, max_new_tokens, temperature=1.0, top_k=None):
+        """
+        Take a conditioning sequence of indices idx (LongTensor of shape (b,t)) and complete
+        the sequence max_new_tokens times, feeding the predictions back into the model each time.
+        Most likely you'll want to make sure to be in model.eval() mode of operation for this.
+        """
+        for _ in range(max_new_tokens):
+            # if the sequence context is growing too long we must crop it at block_size
+            idx_cond = idx if idx.size(1) <= self.config.block_size else idx[:, -self.config.block_size:]
+            # forward the model to get the logits for the index in the sequence
+            logits, _ = self(idx_cond)
+            # pluck the logits at the final step and scale by desired temperature
+            logits = logits[:, -1, :] / temperature
+            # optionally crop the logits to only the top k options
+            if top_k is not None:
+                v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
+                logits[logits < v[:, [-1]]] = -float('Inf')
+            # apply softmax to convert logits to (normalized) probabilities
+            probs = F.softmax(logits, dim=-1)
+            # sample from the distribution
+            idx_next = torch.multinomial(probs, num_samples=1)
+            # append sampled index to the running sequence and continue
+            idx = torch.cat((idx, idx_next), dim=1)
+
+        return idx
+
+

saved_model/ckpt.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d605ac98cb44df934630528b0b8b37ccabcb3f11034c6762928211fd8edf83f
+oid sha256:22dd1a5c7d1d2040dd8d15df01c75f69de6cbaba8aff362718cdd4b4e7fa05ff
 size 1544198298