Upload NGramDebertaV2ForMaskedLM

README.md

@@ -0,0 +1,199 @@
+---
+library_name: transformers
+tags: []
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]

config.json

@@ -0,0 +1,45 @@
+{
+  "architectures": [
+    "NGramDebertaV2ForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "auto_map": {
+    "AutoConfig": "ngram_model.NGramConfigMLM",
+    "AutoModel": "ngram_model.NGramDebertaV2Model",
+    "AutoModelForMaskedLM": "ngram_model.NGramDebertaV2ForMaskedLM"
+  },
+  "bos_token_id": 1,
+  "cls_token_id": 1,
+  "dropout": 0.1,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1280,
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 1024,
+  "max_relative_positions": -1,
+  "model_type": "ngram-deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "sep_token_id": 2,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.54.1",
+  "type_vocab_size": 0,
+  "vocab_size": 40000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:168fe2853ca2ca4f1b0cfc3038b8aa5d6139216f78003581b49899121d9cdc4d
+size 200181560

ngram_model.py

@@ -0,0 +1,101 @@
+from transformers import DebertaV2Model, DebertaV2Config, DebertaV2ForMaskedLM, AutoTokenizer
+from transformers.modeling_outputs import BaseModelOutput
+from typing import Optional, Union
+import torch
+import torch.nn as nn
+
+
+def get_all_substrings(s, min_size=1, max_size=None):
+    substrings = []
+    n = len(s) if max_size is None else min(len(s), max_size)
+    for i in range(min_size-1, n):
+        for j in range(i + 1, n + 1):
+            substrings.append(s[i:j])
+    return substrings
+
+class NGramEmbeds(nn.Module):
+    def __init__(self, config, tokenizer=None):
+        super().__init__()
+        self.config = config
+        self.tokenizer = tokenizer
+        self.ref_table = self.prepare_vocab_table()
+        self.projection = nn.Linear(self.config.vocab_size, self.config.hidden_size)
+
+    def prepare_vocab_table(self):
+        table = torch.eye(self.config.vocab_size, dtype=torch.float)
+        for i in range(self.config.vocab_size):
+            token = self.tokenizer.convert_ids_to_tokens(i)
+
+            if i < 5: # special tokens
+                continue
+            substrings = get_all_substrings(token)  
+            substring_ids = self.tokenizer.convert_tokens_to_ids(substrings)
+
+            table[i, torch.tensor(substring_ids, dtype=torch.long)] = 1.0
+
+        return table
+        # self.register_buffer('ref_table', table) # move to cuda
+
+    def forward(self, input_ids):
+        self.ref_table = self.ref_table.to(input_ids.device)
+        all_ids = self.ref_table[input_ids] # batch_size, seq_len, vocab_size
+        all_embs = self.projection(all_ids)
+        return all_embs
+
+class NGramConfigMLM(DebertaV2Config):
+    model_type = "ngram-deberta-v2"
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+class NGramDebertaV2Model(DebertaV2Model):
+    config_class = NGramConfigMLM
+    def __init__(self, config, tokenizer=None):
+        super().__init__(config)
+        if tokenizer is not None:
+            self.tokenizer = tokenizer
+        else:
+            self.tokenizer = AutoTokenizer.from_pretrained(config._name_or_path)
+
+        self.NGram_embeddings = NGramEmbeds(config, self.tokenizer)
+        self.layer_norm = nn.LayerNorm(config.hidden_size, config.layer_norm_eps)
+        self.post_init()
+    
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[tuple, BaseModelOutput]:
+        word_embeds = self.layer_norm(self.embeddings.word_embeddings(input_ids))
+        NGram_embeds = self.layer_norm(self.NGram_embeddings(input_ids))
+        # embeds = self.embeddings(inputs_embeds=word_embeds + NGram_embeds)
+        embeds = self.embeddings(inputs_embeds=word_embeds + NGram_embeds)
+        return super().forward(
+            attention_mask=attention_mask, 
+            token_type_ids=token_type_ids, 
+            position_ids=position_ids, 
+            inputs_embeds=embeds, 
+            output_attentions=output_attentions, 
+            output_hidden_states=output_hidden_states, 
+            return_dict=return_dict)
+        
+
+
+class NGramDebertaV2ForMaskedLM(DebertaV2ForMaskedLM):
+    config_class = NGramConfigMLM
+    def __init__(self, config, tokenizer=None):
+        super().__init__(config)
+        if tokenizer is not None:
+            self.tokenizer = tokenizer
+        else:
+            self.tokenizer = AutoTokenizer.from_pretrained(config._name_or_path)
+
+        self.deberta = NGramDebertaV2Model(config, self.tokenizer)
+        self.post_init()
+