Model save

README.md

@@ -0,0 +1,53 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: NeuML/bert-hash-nano
+tags:
+- generated_from_trainer
+model-index:
+- name: without_distillation
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# without_distillation
+
+This model is a fine-tuned version of [NeuML/bert-hash-nano](https://huggingface.co/NeuML/bert-hash-nano) on an unknown dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0005
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 10
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 5.8.1
+- Pytorch 2.10.0+cu128
+- Datasets 4.8.3
+- Tokenizers 0.22.2

config.json

@@ -0,0 +1,47 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "BertHashForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "auto_map": {
+    "AutoConfig": "configuration_bert_hash.BertHashConfig",
+    "AutoModel": "modeling_bert_hash.BertHashModel",
+    "AutoModelForMaskedLM": "modeling_bert_hash.BertHashForMaskedLM",
+    "AutoModelForSequenceClassification": "modeling_bert_hash.BertHashForSequenceClassification"
+  },
+  "bos_token_id": null,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 128,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 512,
+  "is_decoder": false,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert_hash",
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "projections": 16,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.8.1",
+  "type_vocab_size": 2,
+  "use_cache": false,
+  "vocab_size": 30522
+}

configuration_bert_hash.py

@@ -0,0 +1,14 @@
+from transformers.models.bert.configuration_bert import BertConfig
+
+
+class BertHashConfig(BertConfig):
+    """
+    Extension of Bert configuration to add projections parameter.
+    """
+
+    model_type = "bert_hash"
+
+    def __init__(self, projections=5, **kwargs):
+        super().__init__(**kwargs)
+
+        self.projections = projections

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4000eb0c9c0a0051512c218a5a4fa3a21d6004df298de68d257294dfdb9fc833
+size 3884844

modeling_bert_hash.py

@@ -0,0 +1,432 @@
+from typing import Optional, Union
+
+import torch
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from transformers.cache_utils import Cache, DynamicCache, EncoderDecoderCache
+from transformers.masking_utils import create_bidirectional_mask, create_causal_mask
+from transformers.models.bert.modeling_bert import BertEncoder, BertPooler, BertPreTrainedModel, BertOnlyMLMHead
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPoolingAndCrossAttentions,
+    MaskedLMOutput,
+    SequenceClassifierOutput,
+)
+from transformers.processing_utils import Unpack
+from transformers.utils import TransformersKwargs, auto_docstring, logging
+from transformers.utils.generic import can_return_tuple, merge_with_config_defaults
+from transformers.utils.output_capturing import capture_outputs
+
+from .configuration_bert_hash import BertHashConfig
+
+logger = logging.get_logger(__name__)
+
+
+class BertHashTokens(nn.Module):
+    """
+    Module that embeds token vocabulary to an intermediate embeddings layer then projects those embeddings to the
+    hidden size.
+
+    The number of projections is like a hash. Setting the projections parameter to 5 is like generating a
+    160-bit hash (5 x float32) for each token. That hash is then projected to the hidden size.
+
+    This significantly reduces the number of parameters necessary for token embeddings.
+
+    For example:
+        Standard token embeddings:
+          30,522 (vocab size) x 768 (hidden size) = 23,440,896 parameters
+          23,440,896 x 4 (float32) = 93,763,584 bytes
+
+        Hash token embeddings:
+          30,522 (vocab size) x 5 (hash buckets) + 5 x 768 (projection matrix)= 156,450 parameters
+          156,450 x 4 (float32) = 625,800 bytes    
+    """
+
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+
+        # Token embeddings
+        self.embeddings = nn.Embedding(config.vocab_size, config.projections, padding_idx=config.pad_token_id)
+
+        # Token embeddings projections
+        self.projections = nn.Linear(config.projections, config.hidden_size)
+
+    def forward(self, input_ids):
+        # Project embeddings to hidden size
+        return self.projections(self.embeddings(input_ids))
+
+
+class BertHashEmbeddings(nn.Module):
+    """Construct the embeddings from word, position and token_type embeddings."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.word_embeddings = BertHashTokens(config)
+        self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
+        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
+
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+        self.register_buffer(
+            "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)), persistent=False
+        )
+        self.register_buffer(
+            "token_type_ids", torch.zeros(self.position_ids.size(), dtype=torch.long), persistent=False
+        )
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor | None = None,
+        token_type_ids: torch.LongTensor | None = None,
+        position_ids: torch.LongTensor | None = None,
+        inputs_embeds: torch.FloatTensor | None = None,
+        past_key_values_length: int = 0,
+    ) -> torch.Tensor:
+        if input_ids is not None:
+            input_shape = input_ids.size()
+        else:
+            input_shape = inputs_embeds.size()[:-1]
+
+        batch_size, seq_length = input_shape
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if position_ids is None:
+            position_ids = (
+                torch.arange(seq_length, dtype=torch.long, device=device)
+                .unsqueeze(0)
+                .expand(batch_size, seq_length)
+            )
+
+        # Setting the token_type_ids to the registered buffer in constructor where it is all zeros, which usually occurs
+        # when its auto-generated, registered buffer helps users when tracing the model without passing token_type_ids, solves
+        # issue #5664
+        if token_type_ids is None:
+            if hasattr(self, "token_type_ids"):
+                # NOTE: We assume either pos ids to have bsz == 1 (broadcastable) or bsz == effective bsz (input_shape[0])
+                buffered_token_type_ids = self.token_type_ids.expand(position_ids.shape[0], -1)
+                buffered_token_type_ids = torch.gather(buffered_token_type_ids, dim=1, index=position_ids)
+                token_type_ids = buffered_token_type_ids.expand(batch_size, seq_length)
+            else:
+                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=self.position_ids.device)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.word_embeddings(input_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+        embeddings = inputs_embeds + token_type_embeddings
+
+        position_embeddings = self.position_embeddings(position_ids)
+        embeddings = embeddings + position_embeddings
+
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+
+@auto_docstring(
+    custom_intro="""
+    The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of
+    cross-attention is added between the self-attention layers, following the architecture described in [Attention is
+    all you need](https://huggingface.co/papers/1706.03762) by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,
+    Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
+
+    To behave as an decoder the model needs to be initialized with the `is_decoder` argument of the configuration set
+    to `True`. To be used in a Seq2Seq model, the model needs to initialized with both `is_decoder` argument and
+    `add_cross_attention` set to `True`; an `encoder_hidden_states` is then expected as an input to the forward pass.
+    """
+)
+class BertHashModel(BertPreTrainedModel):
+    config_class = BertHashConfig
+
+    _no_split_modules = ["BertEmbeddings", "BertLayer"]
+
+    def __init__(self, config, add_pooling_layer=True):
+        r"""
+        add_pooling_layer (bool, *optional*, defaults to `True`):
+            Whether to add a pooling layer
+        """
+        super().__init__(config)
+        self.config = config
+        self.gradient_checkpointing = False
+
+        self.embeddings = BertHashEmbeddings(config)
+        self.encoder = BertEncoder(config)
+
+        self.pooler = BertPooler(config) if add_pooling_layer else None
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings.embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    @merge_with_config_defaults
+    @capture_outputs
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: torch.Tensor | None = None,
+        attention_mask: torch.Tensor | None = None,
+        token_type_ids: torch.Tensor | None = None,
+        position_ids: torch.Tensor | None = None,
+        inputs_embeds: torch.Tensor | None = None,
+        encoder_hidden_states: torch.Tensor | None = None,
+        encoder_attention_mask: torch.Tensor | None = None,
+        past_key_values: Cache | None = None,
+        use_cache: bool | None = None,
+        **kwargs: Unpack[TransformersKwargs],
+    ) -> tuple[torch.Tensor] | BaseModelOutputWithPoolingAndCrossAttentions:
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        if self.config.is_decoder:
+            use_cache = use_cache if use_cache is not None else self.config.use_cache
+        else:
+            use_cache = False
+
+        if use_cache and past_key_values is None:
+            past_key_values = (
+                EncoderDecoderCache(DynamicCache(config=self.config), DynamicCache(config=self.config))
+                if encoder_hidden_states is not None or self.config.is_encoder_decoder
+                else DynamicCache(config=self.config)
+            )
+
+        past_key_values_length = past_key_values.get_seq_length() if past_key_values is not None else 0
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            token_type_ids=token_type_ids,
+            inputs_embeds=inputs_embeds,
+            past_key_values_length=past_key_values_length,
+        )
+
+        attention_mask, encoder_attention_mask = self._create_attention_masks(
+            attention_mask=attention_mask,
+            encoder_attention_mask=encoder_attention_mask,
+            embedding_output=embedding_output,
+            encoder_hidden_states=encoder_hidden_states,
+            past_key_values=past_key_values,
+        )
+
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=attention_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            past_key_values=past_key_values,
+            use_cache=use_cache,
+            position_ids=position_ids,
+            **kwargs,
+        )
+        sequence_output = encoder_outputs.last_hidden_state
+        pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
+
+        return BaseModelOutputWithPoolingAndCrossAttentions(
+            last_hidden_state=sequence_output,
+            pooler_output=pooled_output,
+            past_key_values=encoder_outputs.past_key_values,
+        )
+
+    def _create_attention_masks(
+        self,
+        attention_mask,
+        encoder_attention_mask,
+        embedding_output,
+        encoder_hidden_states,
+        past_key_values,
+    ):
+        if self.config.is_decoder:
+            attention_mask = create_causal_mask(
+                config=self.config,
+                inputs_embeds=embedding_output,
+                attention_mask=attention_mask,
+                past_key_values=past_key_values,
+            )
+        else:
+            attention_mask = create_bidirectional_mask(
+                config=self.config,
+                inputs_embeds=embedding_output,
+                attention_mask=attention_mask,
+            )
+
+        if encoder_attention_mask is not None:
+            encoder_attention_mask = create_bidirectional_mask(
+                config=self.config,
+                inputs_embeds=embedding_output,
+                attention_mask=encoder_attention_mask,
+                encoder_hidden_states=encoder_hidden_states,
+            )
+
+        return attention_mask, encoder_attention_mask
+
+
+@auto_docstring
+class BertForMaskedLM(BertPreTrainedModel):
+    _tied_weights_keys = {
+        "cls.predictions.decoder.weight": "bert.embeddings.word_embeddings.weight",
+        "cls.predictions.decoder.bias": "cls.predictions.bias",
+    }
+    config_class = BertHashConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        if config.is_decoder:
+            logger.warning(
+                "If you want to use `BertForMaskedLM` make sure `config.is_decoder=False` for "
+                "bi-directional self-attention."
+            )
+
+        self.bert = BertHashModel(config, add_pooling_layer=False)
+        self.cls = BertOnlyMLMHead(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_output_embeddings(self):
+        return self.cls.predictions.decoder
+
+    def set_output_embeddings(self, new_embeddings):
+        self.cls.predictions.decoder = new_embeddings
+        self.cls.predictions.bias = new_embeddings.bias
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: torch.Tensor | None = None,
+        attention_mask: torch.Tensor | None = None,
+        token_type_ids: torch.Tensor | None = None,
+        position_ids: torch.Tensor | None = None,
+        inputs_embeds: torch.Tensor | None = None,
+        encoder_hidden_states: torch.Tensor | None = None,
+        encoder_attention_mask: torch.Tensor | None = None,
+        labels: torch.Tensor | None = None,
+        **kwargs: Unpack[TransformersKwargs],
+    ) -> tuple[torch.Tensor] | MaskedLMOutput:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+        """
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            return_dict=True,
+            **kwargs,
+        )
+
+        sequence_output = outputs[0]
+        prediction_scores = self.cls(sequence_output)
+
+        masked_lm_loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()  # -100 index = padding token
+            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+        return MaskedLMOutput(
+            loss=masked_lm_loss,
+            logits=prediction_scores,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+
+@auto_docstring(
+    custom_intro="""
+    Bert Model transformer with a sequence classification/regression head on top (a linear layer on top of the pooled
+    output) e.g. for GLUE tasks.
+    """
+)
+class BertHashForSequenceClassification(BertPreTrainedModel):
+    config_class = BertHashConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.config = config
+
+        self.bert = BertHashModel(config)
+        classifier_dropout = (
+            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+        )
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: torch.Tensor | None = None,
+        attention_mask: torch.Tensor | None = None,
+        token_type_ids: torch.Tensor | None = None,
+        position_ids: torch.Tensor | None = None,
+        inputs_embeds: torch.Tensor | None = None,
+        labels: torch.Tensor | None = None,
+        **kwargs: Unpack[TransformersKwargs],
+    ) -> tuple[torch.Tensor] | SequenceClassifierOutput:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            return_dict=True,
+            **kwargs,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits, labels)
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d64176ded3e893eb3173645129081f03bc47574987ac198ce04e75f938f046d
+size 5265