Upload TTMiniLM

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 +1,13 @@
+{
+  "architectures": [
+    "TTMiniLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "modeling_tt.TTConfig",
+    "AutoModel": "modeling_tt.TTMiniLM"
+  },
+  "dtype": "float32",
+  "model_type": "tt_minilm",
+  "rank": 24,
+  "transformers_version": "5.0.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d26c27cc64f113cab958b835e554a4386c82a8f2bd60aa7340149a1768363b15
+size 8331808

modeling_tt.py

@@ -0,0 +1,95 @@
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from transformers import PretrainedConfig, PreTrainedModel, BertConfig, BertModel
+
+class TTConfig(PretrainedConfig):
+    model_type = "tt_minilm"
+    def __init__(self, rank=24, **kwargs):
+        super().__init__(**kwargs)
+        self.rank = rank
+
+class TTLinear(nn.Module):
+    def __init__(self, in_features, out_features, shapes, rank, bias=True):
+        super().__init__()
+        self.in_features = in_features
+        self.out_features = out_features
+        self.shapes = shapes
+        self.rank = rank
+        self.cores = nn.ParameterList()
+        for i, (d_in, d_out) in enumerate(shapes):
+            r_in = 1 if i == 0 else rank
+            r_out = 1 if i == len(shapes) - 1 else rank
+            self.cores.append(nn.Parameter(torch.zeros(r_in, d_in, d_out, r_out)))
+        if bias:
+            self.bias = nn.Parameter(torch.zeros(out_features))
+        else:
+            self.register_parameter('bias', None)
+
+    def forward(self, x):
+        res = self.cores[0]
+        for i in range(1, len(self.cores)):
+            res = torch.tensordot(res, self.cores[i], dims=([-1], [0]))
+        res = res.squeeze(0).squeeze(-1)
+        n = len(self.cores)
+        permute_idx = [2 * i for i in range(n)] + [2 * i + 1 for i in range(n)]
+        res = res.permute(*permute_idx).contiguous()
+        weight_matrix = res.view(self.in_features, self.out_features)
+        out = torch.matmul(x, weight_matrix)
+        if self.bias is not None:
+            out += self.bias
+        return out
+
+class TTEmbedding(nn.Module):
+    def __init__(self, num_embeddings, embedding_dim, shapes, rank):
+        super().__init__()
+        self.num_embeddings = num_embeddings
+        self.embedding_dim = embedding_dim
+        self.shapes = shapes
+        self.cores = nn.ParameterList()
+        for i, (d_v, d_e) in enumerate(shapes):
+            r_in = 1 if i == 0 else rank
+            r_out = 1 if i == len(shapes) - 1 else rank
+            self.cores.append(nn.Parameter(torch.zeros(r_in, d_v, d_e, r_out)))
+
+    def forward(self, input_ids):
+        res = self.cores[0]
+        for i in range(1, len(self.cores)):
+            res = torch.tensordot(res, self.cores[i], dims=([-1], [0]))
+        res = res.squeeze(0).squeeze(-1)
+        n = len(self.cores)
+        permute_idx = [2 * i for i in range(n)] + [2 * i + 1 for i in range(n)]
+        weights = res.permute(*permute_idx).contiguous().view(-1, self.embedding_dim)
+        weights = weights[:self.num_embeddings, :]
+        return F.embedding(input_ids, weights)
+
+def compress_to_10_percent_empty(model, rank=24):
+    for name, module in model.named_modules():
+        if isinstance(module, nn.Linear):
+            bias = module.bias is not None
+            if module.in_features == 384 and module.out_features == 384:
+                new_layer = TTLinear(384, 384, [(8, 8), (8, 8), (6, 6)], rank=rank, bias=bias)
+            elif module.in_features == 384 and module.out_features == 1536:
+                new_layer = TTLinear(384, 1536, [(8, 16), (8, 12), (6, 8)], rank=rank, bias=bias)
+            elif module.in_features == 1536 and module.out_features == 384:
+                new_layer = TTLinear(1536, 384, [(16, 8), (12, 8), (8, 6)], rank=rank, bias=bias)
+            else: 
+                continue
+            parent_path = name.rsplit('.', 1)
+            parent = model if len(parent_path) == 1 else dict(model.named_modules())[parent_path[0]]
+            setattr(parent, parent_path[-1], new_layer)
+
+    emb_shapes = [(32, 8), (31, 8), (31, 6)]
+    model.embeddings.word_embeddings = TTEmbedding(30522, 384, emb_shapes, rank=rank)
+
+class TTMiniLM(PreTrainedModel):
+    config_class = TTConfig
+    def __init__(self, config):
+        super().__init__(config)
+        base_config = BertConfig.from_pretrained('sentence-transformers/all-MiniLM-L6-v2')
+        self.bert = BertModel(base_config)
+        compress_to_10_percent_empty(self.bert, rank=config.rank)
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, **kwargs):
+        return self.bert(input_ids=input_ids, attention_mask=attention_mask, token_type_ids=token_type_ids, **kwargs)