Remove unnecessary code

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 +1,24 @@
+{
+  "architectures": [
+    "MELPEncoderModel"
+  ],
+  "attn_pooler_heads": 8,
+  "auto_map": {
+    "AutoConfig": "configuration_MELP_Encoder.MELPEncoderConfig",
+    "AutoModel": "modeling_MELP_Encoder.MELPEncoderModel"
+  },
+  "drop": 0.0,
+  "embed_dim_caption": 768,
+  "model_size": "small",
+  "model_type": "melp",
+  "n_queries_caption": 128,
+  "n_queries_contrast": 14,
+  "num_leads": 12,
+  "proj": "linear",
+  "proj_bias": false,
+  "shared_emb_dim": 256,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_attentional_pool_caption": true,
+  "use_attentional_pool_contrast": true
+}

configuration_MELP_Encoder.py

@@ -0,0 +1,35 @@
+from transformers import PretrainedConfig
+
+
+class MELPEncoderConfig(PretrainedConfig):
+    model_type = "melp"
+
+    def __init__(
+        self,
+        model_size: str = "small", # small by default
+        shared_emb_dim: int = 256,
+        embed_dim_caption: int = 768,
+        use_attentional_pool_contrast: bool = True,
+        use_attentional_pool_caption: bool = True,
+        n_queries_contrast: int = 14,
+        n_queries_caption: int = 128,
+        attn_pooler_heads: int = 8,
+        proj: str = "linear",
+        drop: float = 0.,
+        proj_bias: bool = False,
+        num_leads: int = 12,
+        **kwargs
+    ):
+        self.model_size = model_size
+        self.shared_emb_dim = shared_emb_dim
+        self.embed_dim_caption = embed_dim_caption
+        self.use_attentional_pool_contrast = use_attentional_pool_contrast
+        self.use_attentional_pool_caption = use_attentional_pool_caption
+        self.n_queries_contrast = n_queries_contrast
+        self.n_queries_caption = n_queries_caption
+        self.attn_pooler_heads = attn_pooler_heads
+        self.proj = proj
+        self.drop = drop
+        self.proj_bias = proj_bias
+        self.num_leads = num_leads
+        super().__init__(**kwargs)

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebd02d76e12c9740c32d596f3341fc04f4b6277d2bb7483302b47533dd00f10f
+size 131213666

modeling_MELP_Encoder.py

@@ -0,0 +1,245 @@
+import torch
+import numpy as np
+import torch.nn as nn
+import torch.nn.functional as F
+from collections import OrderedDict
+from typing import Callable
+from timm.models.layers import Mlp
+from fairseq_signals_backbone.models.wav2vec2.wav2vec2_cmsc import Wav2Vec2CMSCModel, Wav2Vec2CMSCConfig
+from lightning import LightningModule
+from transformers import PreTrainedModel
+from .configuration_MELP_Encoder import MELPEncoderConfig
+
+
+class LayerNorm(nn.LayerNorm):
+    """Subclass torch's LayerNorm (with cast back to input dtype)."""
+
+    def forward(self, x: torch.Tensor):
+        orig_type = x.dtype
+        x = F.layer_norm(x, self.normalized_shape, self.weight, self.bias, self.eps)
+        return x.to(orig_type)
+    
+
+class AttentionalPooler(nn.Module):
+    def __init__(
+            self,
+            d_model: int,
+            context_dim: int,
+            n_head: int = 8,
+            n_queries: int = 256,
+            norm_layer: Callable = LayerNorm,
+    ):
+        super().__init__()
+        self.query = nn.Parameter(torch.randn(n_queries, d_model))
+        self.attn = nn.MultiheadAttention(d_model, n_head, kdim=context_dim, vdim=context_dim, batch_first=True)
+        self.ln_q = norm_layer(d_model)
+        self.ln_k = norm_layer(context_dim)
+
+    def forward(self, x: torch.Tensor):
+        N = x.shape[0]
+        x = self.ln_k(x)
+        q = self.ln_q(self.query)
+        out = self.attn(q.unsqueeze(0).expand(N, -1, -1), x, x, need_weights=False)[0]
+        return out
+
+
+def off_diagonal(x):
+    # return a flattened view of the off-diagonal elements of a square matrix
+    n, m = x.shape
+    assert n == m
+    return x.flatten()[:-1].view(n - 1, n + 1)[:, 1:].flatten()
+
+
+class ECGFMModel(LightningModule):
+    def __init__(self, 
+                 model_size: str = "small", # small by default
+                 shared_emb_dim: int = 256,
+                 embed_dim_caption: int = 768,
+                 use_attentional_pool_contrast: bool = False,
+                 use_attentional_pool_caption: bool = False,
+                 n_queries_contrast: int = 10,
+                 n_queries_caption: int = 128,
+                 attn_pooler_heads: int = 8,
+                 norm_layer: nn.Module = nn.LayerNorm,
+                 proj: str = "linear",
+                 drop: float = 0.,
+                 proj_bias: bool = False,
+                 num_leads: int = 12,
+                 softmax_temperature: float = 0.1,
+                 lambd: float = 0.0051,
+                 *args,
+                 **kwargs):
+        
+        """" Implementation of ECG-FM model.
+        Using the Wave2Vec2 model as the ECG encoder: CNN + Transformer
+        
+        """
+        super().__init__()
+        self.save_hyperparameters()
+        self.shared_emb_dim = shared_emb_dim
+        self.num_leads = num_leads
+        self.temperature = softmax_temperature
+
+        if model_size == "small":
+            self.encoder_embed_dim = 768
+            self.encoder_attention_heads = 12
+            self.encoder_layers = 8
+            self.encoder_ffn_embed_dim = 3072
+        elif model_size == "base":
+            self.encoder_embed_dim = 768
+            self.encoder_attention_heads = 12
+            self.encoder_layers = 12
+            self.encoder_ffn_embed_dim = 3072
+        elif model_size == "large":
+            self.encoder_embed_dim = 1024
+            self.encoder_attention_heads = 16
+            self.encoder_layers = 24
+            self.encoder_ffn_embed_dim = 4096
+        else:
+            raise ValueError(f"Unknown model size: {model_size}")
+        print("Using ECG encoder with the following configuration:")
+        print(f"encoder_embed_dim: {self.encoder_embed_dim}")
+        print(f"encoder_attention_heads: {self.encoder_attention_heads}")
+        print(f"encoder_layers: {self.encoder_layers}")
+        print(f"encoder_ffn_embed_dim: {self.encoder_ffn_embed_dim}")
+        
+        self.init_ecg_encoder()
+
+        self.embed_dim_caption = embed_dim_caption
+        self.use_attentional_pool_contrast = use_attentional_pool_contrast
+        self.use_attentional_pool_caption = use_attentional_pool_caption
+    
+        head_layers = OrderedDict()
+        prev_chs = self.ecg_encoder.cfg.encoder_embed_dim
+        if use_attentional_pool_contrast:
+            scale = prev_chs ** -0.5
+            self.attn_pool_contrast = AttentionalPooler(
+                d_model=shared_emb_dim, 
+                context_dim=prev_chs, 
+                n_head=attn_pooler_heads, 
+                n_queries=n_queries_contrast)
+            self.ln_contrast = norm_layer(shared_emb_dim)
+            self.proj_contrast = nn.Parameter(scale * torch.randn(shared_emb_dim, shared_emb_dim))
+        else:
+            assert proj, 'projection layer needed if not using attentional pooling.'
+            # NOTE attention pool ends with a projection layer, so proj should usually be set to '' if such pooling is used
+            if proj == 'linear':
+                head_layers['drop'] = nn.Dropout(drop)
+                head_layers['proj'] = nn.Linear(prev_chs, shared_emb_dim, bias=proj_bias)
+            elif proj == 'mlp':
+                head_layers['mlp'] = Mlp(prev_chs, 2 * shared_emb_dim, shared_emb_dim, drop=(drop, 0), bias=(True, proj_bias))
+
+        self.head = nn.Sequential(head_layers)
+
+        if use_attentional_pool_caption:
+            self.attn_pool_caption = AttentionalPooler(
+                d_model=embed_dim_caption, context_dim=prev_chs, n_head=attn_pooler_heads, n_queries=n_queries_caption)
+            self.ln_caption = norm_layer(embed_dim_caption)
+        
+        self.logit_scale = nn.Parameter(torch.ones([]) * np.log(1 / 0.07))
+
+        self.bn = nn.BatchNorm1d(768, affine=False)
+        self.lambd = lambd
+
+    def init_ecg_encoder(self):
+        # Here we define Wav2Vec2CMSC model as the ECG encoder
+        cfg = Wav2Vec2CMSCConfig(
+            apply_mask = True,
+            mask_prob = 0.65,
+            quantize_targets = True,
+            final_dim = 256,
+            dropout_input = 0.1,
+            dropout_features = 0.1,
+            feature_grad_mult = 0.1,
+            encoder_embed_dim = self.encoder_embed_dim,
+            encoder_attention_heads = self.encoder_attention_heads,
+            in_d = 12,
+            encoder_layers = self.encoder_layers,
+            encoder_ffn_embed_dim = self.encoder_ffn_embed_dim
+        )
+        self.ecg_encoder = Wav2Vec2CMSCModel(cfg)
+
+    def _global_pool(self, x):
+        return torch.mean(x, dim=1)
+        
+    @torch.no_grad()
+    # only used for finetune ...
+    def ext_ecg_emb(self, ecg, normalize=False):
+        assert ecg.dim() == 3, "Input tensor must be 3D"
+
+        ecg_out = self.ecg_encoder(source=ecg, mask=False, features_only=True)
+        features = ecg_out["x"]
+
+        if self.use_attentional_pool_contrast:
+            pooled = self.attn_pool_contrast(features)
+            pooled = self.ln_contrast(pooled)
+            pooled = torch.mean(pooled, dim=1)
+        else:
+            pooled = self._global_pool(features)
+
+        if normalize:
+            pooled = F.normalize(pooled, p=2, dim=-1)
+        
+        return pooled
+
+    def _encode_ecg(self, ecg):
+        assert ecg.dim() == 3, "Input tensor must be 3D"
+        ecg_out = self.ecg_encoder(source=ecg, mask=False, features_only=True)
+        # features = self.ecg_encoder.get_features(net_output=ecg_out, aggregate=False)
+        # results after CNN-Transformer
+        features = ecg_out["x"]
+
+        if self.use_attentional_pool_contrast:
+            # hierarchical pooling
+            pooled = self.attn_pool_contrast(features)
+            pooled = self.ln_contrast(pooled)
+            pooled = pooled @ self.proj_contrast.unsqueeze(0)
+            pooled_beat = pooled.clone()
+            pooled = torch.mean(pooled, dim=1)
+        else:
+            pooled = self._global_pool(features)
+            pooled = self.head(features)
+
+        tokens = None
+        if self.use_attentional_pool_caption:
+            tokens = self.attn_pool_caption(features)
+            tokens = self.ln_caption(tokens)
+        else:
+            tokens = None
+
+        return pooled, pooled_beat, tokens
+    
+    def encode_ecg(self, ecg):
+        ecg_latent, _, _ = self._encode_ecg(ecg)
+        return ecg_latent
+
+
+class MELPEncoderModel(PreTrainedModel):
+    config_class = MELPEncoderConfig
+
+    def __init__(self, config: MELPEncoderConfig):
+        super().__init__(config)
+
+        self.ecg_encoder = ECGFMModel(
+            model_size=config.model_size,
+            shared_emb_dim=config.shared_emb_dim,
+            embed_dim_caption=config.embed_dim_caption,
+            use_attentional_pool_contrast=config.use_attentional_pool_contrast,
+            use_attentional_pool_caption=config.use_attentional_pool_caption,
+            n_queries_contrast=config.n_queries_contrast,
+            n_queries_caption=config.n_queries_caption,
+            attn_pooler_heads=config.attn_pooler_heads,
+            proj=config.proj,
+            drop=config.drop,
+            proj_bias=config.proj_bias,
+            num_leads=config.num_leads,
+        )
+    
+    def forward(self, tensor: torch.Tensor) -> torch.Tensor:
+        proj_ecg_emb, ecg_beat_emb, ecg_token_emb = self.ecg_encoder._encode_ecg(tensor)
+
+        return {
+            "proj_ecg_emb": proj_ecg_emb,
+            "ecg_beat_emb": ecg_beat_emb,
+            "ecg_token_emb": ecg_token_emb
+        }