Run GLUE and remove LAMB

examples/run_glue.py

@@ -47,7 +47,11 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLNetTokenizer,
                                   DistilBertConfig,
                                   DistilBertForSequenceClassification,
-                                  DistilBertTokenizer)
+                                  DistilBertTokenizer,
+                                  AlbertConfig,
+                                  AlbertForSequenceClassification, 
+                                  AlbertTokenizer,
+                                )
 
 from transformers import AdamW, get_linear_schedule_with_warmup
 
@@ -66,7 +70,8 @@ MODEL_CLASSES = {
     'xlnet': (XLNetConfig, XLNetForSequenceClassification, XLNetTokenizer),
     'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
     'roberta': (RobertaConfig, RobertaForSequenceClassification, RobertaTokenizer),
-    'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
+    'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer),
+    'albert': (AlbertConfig, AlbertForSequenceClassification, AlbertTokenizer)
 }
 
 
@@ -99,6 +104,7 @@ def train(args, train_dataset, model, tokenizer):
         {'params': [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)], 'weight_decay': args.weight_decay},
         {'params': [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}
         ]
+
     optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
     scheduler = get_linear_schedule_with_warmup(optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total)
     if args.fp16:
@@ -317,7 +323,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
         all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
     elif output_mode == "regression":
         all_labels = torch.tensor([f.label for f in features], dtype=torch.float)
-
+ 
     dataset = TensorDataset(all_input_ids, all_attention_mask, all_token_type_ids, all_labels)
     return dataset
 
@@ -361,7 +367,7 @@ def main():
     parser.add_argument("--per_gpu_eval_batch_size", default=8, type=int,
                         help="Batch size per GPU/CPU for evaluation.")
     parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
-                        help="Number of updates steps to accumulate before performing a backward/update pass.")
+                        help="Number of updates steps to accumulate before performing a backward/update pass.")     
     parser.add_argument("--learning_rate", default=5e-5, type=float,
                         help="The initial learning rate for Adam.")
     parser.add_argument("--weight_decay", default=0.0, type=float,

transformers/optimization.py

@@ -167,96 +167,3 @@ class AdamW(Optimizer):
                     p.data.add_(-group['lr'] * group['weight_decay'], p.data)
 
         return loss
-
-
-
-class Lamb(Optimizer):
-    """ Implements the LAMB algorithm (Layer-wise Adaptive Moments optimizer for Batch training).
-
-    Adapted from the huggingface/transformers ADAM optimizer
-    Inspired from the Google Research implementation available in ALBERT: https://github.com/google-research/google-research/blob/master/albert/lamb_optimizer.py
-    Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
-
-
-    Parameters:
-        lr (float): learning rate. Default 1e-3.
-        betas (tuple of 2 floats): Adams beta parameters (b1, b2). Default: (0.9, 0.999)
-        eps (float): Adams epsilon. Default: 1e-6
-        weight_decay (float): Weight decay. Default: 0.0
-    """
-    
-    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0.0, correct_bias=True):
-        if lr < 0.0:
-            raise ValueError("Invalid learning rate: {} - should be >= 0.0".format(lr))
-        if not 0.0 <= betas[0] < 1.0:
-            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[0]))
-        if not 0.0 <= betas[1]  < 1.0:
-            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[1]))
-        if not 0.0 <= eps:
-            raise ValueError("Invalid epsilon value: {} - should be >= 0.0".format(eps))
-        defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay,
-                        correct_bias=correct_bias)
-        super(Lamb, self).__init__(params, defaults)
-
-    def step(self, closure=None):
-        """Performs a single optimization step.
-
-        Arguments:
-            closure (callable, optional): A closure that reevaluates the model
-                and returns the loss.
-        """
-        loss = None
-        if closure is not None:
-            loss = closure()
-
-        for group in self.param_groups:
-            for p in group['params']:
-                if p.grad is None:
-                    continue
-                grad = p.grad.data
-                if grad.is_sparse:
-                    raise RuntimeError('LAMB does not support sparse gradients.')
-
-                state = self.state[p]
-
-                # State initialization
-                if len(state) == 0:
-                    state['step'] = 0
-                    # Exponential moving average of gradient values
-                    state['exp_avg'] = torch.zeros_like(p.data)
-                    # Exponential moving average of squared gradient values
-                    state['exp_avg_sq'] = torch.zeros_like(p.data)
-
-                exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
-                beta1, beta2 = group['betas']
-
-                state['step'] += 1
-
-                # Decay the first and second moment running average coefficient
-                # In-place operations to update the averages at the same time
-                exp_avg.mul_(beta1).add_(1.0 - beta1, grad)
-                exp_avg_sq.mul_(beta2).addcmul_(1.0 - beta2, grad, grad)
-                denom = exp_avg_sq.sqrt().add_(group['eps'])
-
-
-                # Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
-                step_size = group['lr']
-                weight_norm = p.data.pow(2).sum().sqrt().clamp(0, 10)
-
-                adam_step = exp_avg / exp_avg_sq.sqrt().add(group['eps'])
-                if group['weight_decay'] != 0:
-                    adam_step.add_(group['weight_decay'], p.data)
-
-                adam_norm = adam_step.pow(2).sum().sqrt()
-                if weight_norm == 0 or adam_norm == 0:
-                    trust_ratio = 1
-                else:
-                    trust_ratio = weight_norm / adam_norm
-
-                
-                state['weight_norm'] = weight_norm
-                state['adam_norm'] = adam_norm
-                state['trust_ratio'] = trust_ratio
-
-                p.data.add_(-step_size * trust_ratio, adam_step)
-        return loss