Upload 69 files

BERT/.ipynb_checkpoints/Untitled-checkpoint.ipynb

@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

BERT/.ipynb_checkpoints/Untitled1-checkpoint.ipynb

@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

BERT/BERT-EXPL.py

@@ -0,0 +1,66 @@
+import os
+from transformers import BertTokenizer
+from BERT_explainability.modules.BERT.ExplanationGenerator import Generator
+from BERT_explainability.modules.BERT.BertForSequenceClassification import BertForSequenceClassification
+from transformers import BertTokenizer
+from BERT_explainability.modules.BERT.ExplanationGenerator import Generator
+from transformers import AutoTokenizer
+from captum.attr import visualization
+import spacy
+import torch
+from IPython.display import Image, HTML, display
+
+from sequenceoutput.modeling_output import SequenceClassifierOutput
+
+model = BertForSequenceClassification.from_pretrained("./BERT").to("cuda")
+model.eval()
+tokenizer = AutoTokenizer.from_pretrained("./BERT")
+# initialize the explanations generator
+explanations = Generator(model)
+
+classifications = ["NEGATIVE", "POSITIVE"]
+
+# encode a sentence
+text_batch = ["I hate that I love you."]
+encoding = tokenizer(text_batch, return_tensors='pt')
+input_ids = encoding['input_ids'].to("cuda")
+attention_mask = encoding['attention_mask'].to("cuda")
+
+# true class is positive - 1
+true_class = 1
+
+# generate an explanation for the input
+target_class = 0
+expl = \
+    explanations.generate_LRP(input_ids=input_ids, attention_mask=attention_mask, start_layer=11, index=target_class)[0]
+# normalize scores
+expl = (expl - expl.min()) / (expl.max() - expl.min())
+
+# get the model classification
+output = torch.nn.functional.softmax(model(input_ids=input_ids, attention_mask=attention_mask)[0], dim=-1)
+classification = output.argmax(dim=-1).item()
+# get class name
+class_name = classifications[target_class]
+# if the classification is negative, higher explanation scores are more negative
+# flip for visualization
+if class_name == "NEGATIVE":
+    expl *= (-1)
+token_importance = {}
+tokens = tokenizer.convert_ids_to_tokens(input_ids.flatten())
+for i in range(len(tokens)):
+    token_importance[tokens[i]] = expl[i].item()
+vis_data_records = [visualization.VisualizationDataRecord(
+    expl,
+    output[0][classification],
+    classification,
+    true_class,
+    true_class,
+    1,
+    tokens,
+    1)]
+
+html1 = visualization.visualize_text(vis_data_records)
+# print(token_importance, html1)
+# with open('bert-xai.html', 'w+') as f:
+#     f.write(str(html1))
+# return token_importance, html1

BERT/BERT/config.json

@@ -0,0 +1,20 @@
+{
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "finetuning_task": "sst-2",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

BERT/BERT/eval_results_sst-2.txt

@@ -0,0 +1,3 @@
+eval_loss = 0.2785584788237299
+eval_acc = 0.9243119266055045
+epoch = 3.0

BERT/BERT/flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57ebdee44ea63b8f3a2a53011dabbd37a7bec8da5d38834beb9751075bb8b821
+size 437942328

BERT/BERT/gitattributes

@@ -0,0 +1,9 @@
+*.bin.* filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tar.gz filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text

BERT/BERT/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5f7d1b5618ba58907379af830ee895c8800e3b381286b13d07d90aaf204dc40
+size 437985387

BERT/BERT/special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

BERT/BERT/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "model_max_length": 512}

BERT/BERT/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da4b38103a827982f36030842c04dcc7f34bb64cb2f56fa45cc69860836ca5d1
+size 1053

BERT/BERT_explainability/modules/BERT/BERT.py

@@ -0,0 +1,671 @@
+from __future__ import absolute_import
+
+import torch
+from torch import nn
+import torch.nn.functional as F
+import math
+from transformers import BertConfig
+from transformers.modeling_outputs import BaseModelOutputWithPooling, BaseModelOutput
+from BERT_explainability.modules.layers_ours import *
+from transformers import (
+    BertPreTrainedModel,
+    PreTrainedModel,
+)
+
+ACT2FN = {
+    "relu": ReLU,
+    "tanh": Tanh,
+    "gelu": GELU,
+}
+
+
+def get_activation(activation_string):
+    if activation_string in ACT2FN:
+        return ACT2FN[activation_string]
+    else:
+        raise KeyError("function {} not found in ACT2FN mapping {}".format(activation_string, list(ACT2FN.keys())))
+
+def compute_rollout_attention(all_layer_matrices, start_layer=0):
+    # adding residual consideration
+    num_tokens = all_layer_matrices[0].shape[1]
+    batch_size = all_layer_matrices[0].shape[0]
+    eye = torch.eye(num_tokens).expand(batch_size, num_tokens, num_tokens).to(all_layer_matrices[0].device)
+    all_layer_matrices = [all_layer_matrices[i] + eye for i in range(len(all_layer_matrices))]
+    all_layer_matrices = [all_layer_matrices[i] / all_layer_matrices[i].sum(dim=-1, keepdim=True)
+                          for i in range(len(all_layer_matrices))]
+    joint_attention = all_layer_matrices[start_layer]
+    for i in range(start_layer+1, len(all_layer_matrices)):
+        joint_attention = all_layer_matrices[i].bmm(joint_attention)
+    return joint_attention
+
+class BertEmbeddings(nn.Module):
+    """Construct the embeddings from word, position and token_type embeddings."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
+        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 is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = 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)))
+
+        self.add1 = Add()
+        self.add2 = Add()
+
+    def forward(self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None):
+        if input_ids is not None:
+            input_shape = input_ids.size()
+        else:
+            input_shape = inputs_embeds.size()[:-1]
+
+        seq_length = input_shape[1]
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, :seq_length]
+
+        if token_type_ids is None:
+            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)
+        position_embeddings = self.position_embeddings(position_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        # embeddings = inputs_embeds + position_embeddings + token_type_embeddings
+        embeddings = self.add1([token_type_embeddings, position_embeddings])
+        embeddings = self.add2([embeddings, inputs_embeds])
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+    def relprop(self, cam, **kwargs):
+        cam = self.dropout.relprop(cam, **kwargs)
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+
+        # [inputs_embeds, position_embeddings, token_type_embeddings]
+        (cam) = self.add2.relprop(cam, **kwargs)
+
+        return cam
+
+class BertEncoder(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.layer = nn.ModuleList([BertLayer(config) for _ in range(config.num_hidden_layers)])
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+            output_hidden_states=False,
+            return_dict=False,
+    ):
+        all_hidden_states = () if output_hidden_states else None
+        all_attentions = () if output_attentions else None
+        for i, layer_module in enumerate(self.layer):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer_head_mask = head_mask[i] if head_mask is not None else None
+
+            if getattr(self.config, "gradient_checkpointing", False):
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        return module(*inputs, output_attentions)
+
+                    return custom_forward
+
+                layer_outputs = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(layer_module),
+                    hidden_states,
+                    attention_mask,
+                    layer_head_mask,
+                )
+            else:
+                layer_outputs = layer_module(
+                    hidden_states,
+                    attention_mask,
+                    layer_head_mask,
+                    output_attentions,
+                )
+            hidden_states = layer_outputs[0]
+            if output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, all_hidden_states, all_attentions] if v is not None)
+        return BaseModelOutput(
+            last_hidden_state=hidden_states, hidden_states=all_hidden_states, attentions=all_attentions
+        )
+
+    def relprop(self, cam, **kwargs):
+        # assuming output_hidden_states is False
+        for layer_module in reversed(self.layer):
+            cam = layer_module.relprop(cam, **kwargs)
+        return cam
+
+# not adding relprop since this is only pooling at the end of the network, does not impact tokens importance
+class BertPooler(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.hidden_size)
+        self.activation = Tanh()
+        self.pool = IndexSelect()
+
+    def forward(self, hidden_states):
+        # We "pool" the model by simply taking the hidden state corresponding
+        # to the first token.
+        self._seq_size = hidden_states.shape[1]
+
+        # first_token_tensor = hidden_states[:, 0]
+        first_token_tensor = self.pool(hidden_states, 1, torch.tensor(0, device=hidden_states.device))
+        first_token_tensor = first_token_tensor.squeeze(1)
+        pooled_output = self.dense(first_token_tensor)
+        pooled_output = self.activation(pooled_output)
+        return pooled_output
+
+    def relprop(self, cam, **kwargs):
+        cam = self.activation.relprop(cam, **kwargs)
+        #print(cam.sum())
+        cam = self.dense.relprop(cam, **kwargs)
+        #print(cam.sum())
+        cam = cam.unsqueeze(1)
+        cam = self.pool.relprop(cam, **kwargs)
+        #print(cam.sum())
+
+        return cam
+
+class BertAttention(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.self = BertSelfAttention(config)
+        self.output = BertSelfOutput(config)
+        self.pruned_heads = set()
+        self.clone = Clone()
+
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        heads, index = find_pruneable_heads_and_indices(
+            heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
+        )
+
+        # Prune linear layers
+        self.self.query = prune_linear_layer(self.self.query, index)
+        self.self.key = prune_linear_layer(self.self.key, index)
+        self.self.value = prune_linear_layer(self.self.value, index)
+        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
+        self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+    ):
+        h1, h2 = self.clone(hidden_states, 2)
+        self_outputs = self.self(
+            h1,
+            attention_mask,
+            head_mask,
+            encoder_hidden_states,
+            encoder_attention_mask,
+            output_attentions,
+        )
+        attention_output = self.output(self_outputs[0], h2)
+        outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        # assuming that we don't ouput the attentions (outputs = (attention_output,)), self_outputs=(context_layer,)
+        (cam1, cam2) = self.output.relprop(cam, **kwargs)
+        #print(cam1.sum(), cam2.sum(), (cam1 + cam2).sum())
+        cam1 = self.self.relprop(cam1, **kwargs)
+        #print(cam1.sum(), cam2.sum(), (cam1 + cam2).sum())
+
+        return self.clone.relprop((cam1, cam2), **kwargs)
+
+class BertSelfAttention(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+            raise ValueError(
+                "The hidden size (%d) is not a multiple of the number of attention "
+                "heads (%d)" % (config.hidden_size, config.num_attention_heads)
+            )
+
+        self.num_attention_heads = config.num_attention_heads
+        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        self.query = Linear(config.hidden_size, self.all_head_size)
+        self.key = Linear(config.hidden_size, self.all_head_size)
+        self.value = Linear(config.hidden_size, self.all_head_size)
+
+        self.dropout = Dropout(config.attention_probs_dropout_prob)
+
+        self.matmul1 = MatMul()
+        self.matmul2 = MatMul()
+        self.softmax = Softmax(dim=-1)
+        self.add = Add()
+        self.mul = Mul()
+        self.head_mask = None
+        self.attention_mask = None
+        self.clone = Clone()
+
+        self.attn_cam = None
+        self.attn = None
+        self.attn_gradients = None
+
+    def get_attn(self):
+        return self.attn
+
+    def save_attn(self, attn):
+        self.attn = attn
+
+    def save_attn_cam(self, cam):
+        self.attn_cam = cam
+
+    def get_attn_cam(self):
+        return self.attn_cam
+
+    def save_attn_gradients(self, attn_gradients):
+        self.attn_gradients = attn_gradients
+
+    def get_attn_gradients(self):
+        return self.attn_gradients
+
+    def transpose_for_scores(self, x):
+        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+        x = x.view(*new_x_shape)
+        return x.permute(0, 2, 1, 3)
+
+    def transpose_for_scores_relprop(self, x):
+        return x.permute(0, 2, 1, 3).flatten(2)
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+    ):
+        self.head_mask = head_mask
+        self.attention_mask = attention_mask
+
+        h1, h2, h3 = self.clone(hidden_states, 3)
+        mixed_query_layer = self.query(h1)
+
+        # If this is instantiated as a cross-attention module, the keys
+        # and values come from an encoder; the attention mask needs to be
+        # such that the encoder's padding tokens are not attended to.
+        if encoder_hidden_states is not None:
+            mixed_key_layer = self.key(encoder_hidden_states)
+            mixed_value_layer = self.value(encoder_hidden_states)
+            attention_mask = encoder_attention_mask
+        else:
+            mixed_key_layer = self.key(h2)
+            mixed_value_layer = self.value(h3)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer)
+        key_layer = self.transpose_for_scores(mixed_key_layer)
+        value_layer = self.transpose_for_scores(mixed_value_layer)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        attention_scores = self.matmul1([query_layer, key_layer.transpose(-1, -2)])
+        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
+            attention_scores = self.add([attention_scores, attention_mask])
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = self.softmax(attention_scores)
+
+        self.save_attn(attention_probs)
+        attention_probs.register_hook(self.save_attn_gradients)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = self.matmul2([attention_probs, value_layer])
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        context_layer = context_layer.view(*new_context_layer_shape)
+
+        outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        # Assume output_attentions == False
+        cam = self.transpose_for_scores(cam)
+
+        # [attention_probs, value_layer]
+        (cam1, cam2) = self.matmul2.relprop(cam, **kwargs)
+        cam1 /= 2
+        cam2 /= 2
+        if self.head_mask is not None:
+            # [attention_probs, head_mask]
+            (cam1, _)= self.mul.relprop(cam1, **kwargs)
+
+
+        self.save_attn_cam(cam1)
+
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+
+        cam1 = self.softmax.relprop(cam1, **kwargs)
+
+        if self.attention_mask is not None:
+            # [attention_scores, attention_mask]
+            (cam1, _) = self.add.relprop(cam1, **kwargs)
+
+        # [query_layer, key_layer.transpose(-1, -2)]
+        (cam1_1, cam1_2) = self.matmul1.relprop(cam1, **kwargs)
+        cam1_1 /= 2
+        cam1_2 /= 2
+
+        # query
+        cam1_1 = self.transpose_for_scores_relprop(cam1_1)
+        cam1_1 = self.query.relprop(cam1_1, **kwargs)
+
+        # key
+        cam1_2 = self.transpose_for_scores_relprop(cam1_2.transpose(-1, -2))
+        cam1_2 = self.key.relprop(cam1_2, **kwargs)
+
+        # value
+        cam2 = self.transpose_for_scores_relprop(cam2)
+        cam2 = self.value.relprop(cam2, **kwargs)
+
+        cam = self.clone.relprop((cam1_1, cam1_2, cam2), **kwargs)
+
+        return cam
+
+
+class BertSelfOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.hidden_size)
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.add = Add()
+
+    def forward(self, hidden_states, input_tensor):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        add = self.add([hidden_states, input_tensor])
+        hidden_states = self.LayerNorm(add)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+        # [hidden_states, input_tensor]
+        (cam1, cam2) = self.add.relprop(cam, **kwargs)
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+        cam1 = self.dense.relprop(cam1, **kwargs)
+
+        return (cam1, cam2)
+
+
+class BertIntermediate(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.intermediate_size)
+        if isinstance(config.hidden_act, str):
+            self.intermediate_act_fn = ACT2FN[config.hidden_act]()
+        else:
+            self.intermediate_act_fn = config.hidden_act
+
+    def forward(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.intermediate_act_fn(hidden_states)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        cam = self.intermediate_act_fn.relprop(cam, **kwargs)  # FIXME only ReLU
+        #print(cam.sum())
+        cam = self.dense.relprop(cam, **kwargs)
+        #print(cam.sum())
+        return cam
+
+
+class BertOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.intermediate_size, config.hidden_size)
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.add = Add()
+
+    def forward(self, hidden_states, input_tensor):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        add = self.add([hidden_states, input_tensor])
+        hidden_states = self.LayerNorm(add)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        # print("in", cam.sum())
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+        #print(cam.sum())
+        # [hidden_states, input_tensor]
+        (cam1, cam2)= self.add.relprop(cam, **kwargs)
+        # print("add", cam1.sum(), cam2.sum(), cam1.sum() + cam2.sum())
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+        #print(cam1.sum())
+        cam1 = self.dense.relprop(cam1, **kwargs)
+        # print("dense", cam1.sum())
+
+        # print("out", cam1.sum() + cam2.sum(), cam1.sum(), cam2.sum())
+        return (cam1, cam2)
+
+
+class BertLayer(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.attention = BertAttention(config)
+        self.intermediate = BertIntermediate(config)
+        self.output = BertOutput(config)
+        self.clone = Clone()
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            output_attentions=False,
+    ):
+        self_attention_outputs = self.attention(
+            hidden_states,
+            attention_mask,
+            head_mask,
+            output_attentions=output_attentions,
+        )
+        attention_output = self_attention_outputs[0]
+        outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights
+
+        ao1, ao2 = self.clone(attention_output, 2)
+        intermediate_output = self.intermediate(ao1)
+        layer_output = self.output(intermediate_output, ao2)
+
+        outputs = (layer_output,) + outputs
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        (cam1, cam2) = self.output.relprop(cam, **kwargs)
+        # print("output", cam1.sum(), cam2.sum(), cam1.sum() + cam2.sum())
+        cam1 = self.intermediate.relprop(cam1, **kwargs)
+        # print("intermediate", cam1.sum())
+        cam = self.clone.relprop((cam1, cam2), **kwargs)
+        # print("clone", cam.sum())
+        cam = self.attention.relprop(cam, **kwargs)
+        # print("attention", cam.sum())
+        return cam
+
+
+class BertModel(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = BertEmbeddings(config)
+        self.encoder = BertEncoder(config)
+        self.pooler = BertPooler(config)
+
+        self.init_weights()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    def forward(
+            self,
+            input_ids=None,
+            attention_mask=None,
+            token_type_ids=None,
+            position_ids=None,
+            head_mask=None,
+            inputs_embeds=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=None,
+            output_hidden_states=None,
+            return_dict=None,
+    ):
+        r"""
+        encoder_hidden_states  (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
+            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention
+            if the model is configured as a decoder.
+        encoder_attention_mask (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Mask to avoid performing attention on the padding token indices of the encoder input. This mask
+            is used in the cross-attention if the model is configured as a decoder.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if attention_mask is None:
+            attention_mask = torch.ones(input_shape, device=device)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape, device)
+
+        # If a 2D or 3D attention mask is provided for the cross-attention
+        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+        if self.config.is_decoder and encoder_hidden_states is not None:
+            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
+            if encoder_attention_mask is None:
+                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
+            encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+        else:
+            encoder_extended_attention_mask = None
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids, position_ids=position_ids, token_type_ids=token_type_ids, inputs_embeds=inputs_embeds
+        )
+
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=extended_attention_mask,
+            head_mask=head_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_extended_attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        sequence_output = encoder_outputs[0]
+        pooled_output = self.pooler(sequence_output)
+
+        if not return_dict:
+            return (sequence_output, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=sequence_output,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+    def relprop(self, cam, **kwargs):
+        cam = self.pooler.relprop(cam, **kwargs)
+        # print("111111111111",cam.sum())
+        cam = self.encoder.relprop(cam, **kwargs)
+        # print("222222222222222", cam.sum())
+        # print("conservation: ", cam.sum())
+        return cam
+
+
+if __name__ == '__main__':
+    class Config:
+      def __init__(self, hidden_size, num_attention_heads, attention_probs_dropout_prob):
+        self.hidden_size = hidden_size
+        self.num_attention_heads = num_attention_heads
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+
+    model = BertSelfAttention(Config(1024, 4, 0.1))
+    x = torch.rand(2, 20, 1024)
+    x.requires_grad_()
+
+    model.eval()
+
+    y = model.forward(x)
+
+    relprop = model.relprop(torch.rand(2, 20, 1024), (torch.rand(2, 20, 1024),))
+
+    print(relprop[1][0].shape)

BERT/BERT_explainability/modules/BERT/BERT_cls_lrp.py

@@ -0,0 +1,202 @@
+from transformers import BertPreTrainedModel
+from transformers.utils import logging
+from BERT_explainability.modules.layers_lrp import *
+from BERT_explainability.modules.BERT.BERT_orig_lrp import BertModel
+from torch.nn import CrossEntropyLoss, MSELoss
+import torch.nn as nn
+from typing import List, Any
+import torch
+from BERT_rationale_benchmark.models.model_utils import PaddedSequence
+
+
+class BertForSequenceClassification(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.bert = BertModel(config)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.classifier = Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+    def forward(
+            self,
+            input_ids=None,
+            attention_mask=None,
+            token_type_ids=None,
+            position_ids=None,
+            head_mask=None,
+            inputs_embeds=None,
+            labels=None,
+            output_attentions=None,
+            output_hidden_states=None,
+            return_dict=None,
+    ):
+        r"""
+        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`):
+            Labels for computing the sequence classification/regression loss.
+            Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
+            If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
+            If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(logits.view(-1), labels.view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+    def relprop(self, cam=None, **kwargs):
+        cam = self.classifier.relprop(cam, **kwargs)
+        cam = self.dropout.relprop(cam, **kwargs)
+        cam = self.bert.relprop(cam, **kwargs)
+        return cam
+
+
+# this is the actual classifier we will be using
+class BertClassifier(nn.Module):
+    """Thin wrapper around BertForSequenceClassification"""
+
+    def __init__(self,
+                 bert_dir: str,
+                 pad_token_id: int,
+                 cls_token_id: int,
+                 sep_token_id: int,
+                 num_labels: int,
+                 max_length: int = 512,
+                 use_half_precision=True):
+        super(BertClassifier, self).__init__()
+        bert = BertForSequenceClassification.from_pretrained(bert_dir, num_labels=num_labels)
+        if use_half_precision:
+            import apex
+            bert = bert.half()
+        self.bert = bert
+        self.pad_token_id = pad_token_id
+        self.cls_token_id = cls_token_id
+        self.sep_token_id = sep_token_id
+        self.max_length = max_length
+
+    def forward(self,
+                query: List[torch.tensor],
+                docids: List[Any],
+                document_batch: List[torch.tensor]):
+        assert len(query) == len(document_batch)
+        print(query)
+        # note about device management:
+        # since distributed training is enabled, the inputs to this module can be on *any* device (preferably cpu, since we wrap and unwrap the module)
+        # we want to keep these params on the input device (assuming CPU) for as long as possible for cheap memory access
+        target_device = next(self.parameters()).device
+        cls_token = torch.tensor([self.cls_token_id]).to(device=document_batch[0].device)
+        sep_token = torch.tensor([self.sep_token_id]).to(device=document_batch[0].device)
+        input_tensors = []
+        position_ids = []
+        for q, d in zip(query, document_batch):
+            if len(q) + len(d) + 2 > self.max_length:
+                d = d[:(self.max_length - len(q) - 2)]
+            input_tensors.append(torch.cat([cls_token, q, sep_token, d]))
+            position_ids.append(torch.tensor(list(range(0, len(q) + 1)) + list(range(0, len(d) + 1))))
+        bert_input = PaddedSequence.autopad(input_tensors, batch_first=True, padding_value=self.pad_token_id,
+                                            device=target_device)
+        positions = PaddedSequence.autopad(position_ids, batch_first=True, padding_value=0, device=target_device)
+        (classes,) = self.bert(bert_input.data,
+                               attention_mask=bert_input.mask(on=0.0, off=float('-inf'), device=target_device),
+                               position_ids=positions.data)
+        assert torch.all(classes == classes)  # for nans
+
+        print(input_tensors[0])
+        print(self.relprop()[0])
+
+        return classes
+
+    def relprop(self, cam=None, **kwargs):
+        return self.bert.relprop(cam, **kwargs)
+
+
+if __name__ == '__main__':
+    from transformers import BertTokenizer
+    import os
+
+    class Config:
+        def __init__(self, hidden_size, num_attention_heads, attention_probs_dropout_prob, num_labels,
+                     hidden_dropout_prob):
+            self.hidden_size = hidden_size
+            self.num_attention_heads = num_attention_heads
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.num_labels = num_labels
+            self.hidden_dropout_prob = hidden_dropout_prob
+
+
+    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
+    x = tokenizer.encode_plus("In this movie the acting is great. The movie is perfect! [sep]",
+                         add_special_tokens=True,
+                         max_length=512,
+                         return_token_type_ids=False,
+                         return_attention_mask=True,
+                         pad_to_max_length=True,
+                         return_tensors='pt',
+                         truncation=True)
+
+    print(x['input_ids'])
+
+    model = BertForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=2)
+    model_save_file = os.path.join('./BERT_explainability/output_bert/movies/classifier/', 'classifier.pt')
+    model.load_state_dict(torch.load(model_save_file))
+
+    # x = torch.randint(100, (2, 20))
+    # x = torch.tensor([[101, 2054, 2003, 1996, 15792, 1997, 2023, 3319, 1029, 102,
+    #                    101, 4079, 102, 101, 6732, 102, 101, 2643, 102, 101,
+    #                    2038, 102, 101, 1037, 102, 101, 2933, 102, 101, 2005,
+    #                    102, 101, 2032, 102, 101, 1010, 102, 101, 1037, 102,
+    #                    101, 3800, 102, 101, 2005, 102, 101, 2010, 102, 101,
+    #                    2166, 102, 101, 1010, 102, 101, 1998, 102, 101, 2010,
+    #                    102, 101, 4650, 102, 101, 1010, 102, 101, 2002, 102,
+    #                    101, 2074, 102, 101, 2515, 102, 101, 1050, 102, 101,
+    #                    1005, 102, 101, 1056, 102, 101, 2113, 102, 101, 2054,
+    #                    102, 101, 1012, 102]])
+    # x.requires_grad_()
+
+    model.eval()
+
+    y = model(x['input_ids'], x['attention_mask'])
+    print(y)
+
+    cam, _ = model.relprop()
+
+    #print(cam.shape)
+
+    cam = cam.sum(-1)
+    #print(cam)

BERT/BERT_explainability/modules/BERT/BERT_orig_lrp.py

@@ -0,0 +1,671 @@
+from __future__ import absolute_import
+
+import torch
+from torch import nn
+import torch.nn.functional as F
+import math
+from transformers import BertConfig
+from transformers.modeling_outputs import BaseModelOutputWithPooling, BaseModelOutput
+from BERT_explainability.modules.layers_lrp import *
+from transformers import (
+    BertPreTrainedModel,
+    PreTrainedModel,
+)
+
+ACT2FN = {
+    "relu": ReLU,
+    "tanh": Tanh,
+    "gelu": GELU,
+}
+
+
+def get_activation(activation_string):
+    if activation_string in ACT2FN:
+        return ACT2FN[activation_string]
+    else:
+        raise KeyError("function {} not found in ACT2FN mapping {}".format(activation_string, list(ACT2FN.keys())))
+
+def compute_rollout_attention(all_layer_matrices, start_layer=0):
+    # adding residual consideration
+    num_tokens = all_layer_matrices[0].shape[1]
+    batch_size = all_layer_matrices[0].shape[0]
+    eye = torch.eye(num_tokens).expand(batch_size, num_tokens, num_tokens).to(all_layer_matrices[0].device)
+    all_layer_matrices = [all_layer_matrices[i] + eye for i in range(len(all_layer_matrices))]
+    all_layer_matrices = [all_layer_matrices[i] / all_layer_matrices[i].sum(dim=-1, keepdim=True)
+                          for i in range(len(all_layer_matrices))]
+    joint_attention = all_layer_matrices[start_layer]
+    for i in range(start_layer+1, len(all_layer_matrices)):
+        joint_attention = all_layer_matrices[i].bmm(joint_attention)
+    return joint_attention
+
+class BertEmbeddings(nn.Module):
+    """Construct the embeddings from word, position and token_type embeddings."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
+        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 is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = 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)))
+
+        self.add1 = Add()
+        self.add2 = Add()
+
+    def forward(self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None):
+        if input_ids is not None:
+            input_shape = input_ids.size()
+        else:
+            input_shape = inputs_embeds.size()[:-1]
+
+        seq_length = input_shape[1]
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, :seq_length]
+
+        if token_type_ids is None:
+            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)
+        position_embeddings = self.position_embeddings(position_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        # embeddings = inputs_embeds + position_embeddings + token_type_embeddings
+        embeddings = self.add1([token_type_embeddings, position_embeddings])
+        embeddings = self.add2([embeddings, inputs_embeds])
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+    def relprop(self, cam, **kwargs):
+        cam = self.dropout.relprop(cam, **kwargs)
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+
+        # [inputs_embeds, position_embeddings, token_type_embeddings]
+        (cam) = self.add2.relprop(cam, **kwargs)
+
+        return cam
+
+class BertEncoder(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.layer = nn.ModuleList([BertLayer(config) for _ in range(config.num_hidden_layers)])
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+            output_hidden_states=False,
+            return_dict=False,
+    ):
+        all_hidden_states = () if output_hidden_states else None
+        all_attentions = () if output_attentions else None
+        for i, layer_module in enumerate(self.layer):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer_head_mask = head_mask[i] if head_mask is not None else None
+
+            if getattr(self.config, "gradient_checkpointing", False):
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        return module(*inputs, output_attentions)
+
+                    return custom_forward
+
+                layer_outputs = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(layer_module),
+                    hidden_states,
+                    attention_mask,
+                    layer_head_mask,
+                )
+            else:
+                layer_outputs = layer_module(
+                    hidden_states,
+                    attention_mask,
+                    layer_head_mask,
+                    output_attentions,
+                )
+            hidden_states = layer_outputs[0]
+            if output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, all_hidden_states, all_attentions] if v is not None)
+        return BaseModelOutput(
+            last_hidden_state=hidden_states, hidden_states=all_hidden_states, attentions=all_attentions
+        )
+
+    def relprop(self, cam, **kwargs):
+        # assuming output_hidden_states is False
+        for layer_module in reversed(self.layer):
+            cam = layer_module.relprop(cam, **kwargs)
+        return cam
+
+# not adding relprop since this is only pooling at the end of the network, does not impact tokens importance
+class BertPooler(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.hidden_size)
+        self.activation = Tanh()
+        self.pool = IndexSelect()
+
+    def forward(self, hidden_states):
+        # We "pool" the model by simply taking the hidden state corresponding
+        # to the first token.
+        self._seq_size = hidden_states.shape[1]
+
+        # first_token_tensor = hidden_states[:, 0]
+        first_token_tensor = self.pool(hidden_states, 1, torch.tensor(0, device=hidden_states.device))
+        first_token_tensor = first_token_tensor.squeeze(1)
+        pooled_output = self.dense(first_token_tensor)
+        pooled_output = self.activation(pooled_output)
+        return pooled_output
+
+    def relprop(self, cam, **kwargs):
+        cam = self.activation.relprop(cam, **kwargs)
+        #print(cam.sum())
+        cam = self.dense.relprop(cam, **kwargs)
+        #print(cam.sum())
+        cam = cam.unsqueeze(1)
+        cam = self.pool.relprop(cam, **kwargs)
+        #print(cam.sum())
+
+        return cam
+
+class BertAttention(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.self = BertSelfAttention(config)
+        self.output = BertSelfOutput(config)
+        self.pruned_heads = set()
+        self.clone = Clone()
+
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        heads, index = find_pruneable_heads_and_indices(
+            heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
+        )
+
+        # Prune linear layers
+        self.self.query = prune_linear_layer(self.self.query, index)
+        self.self.key = prune_linear_layer(self.self.key, index)
+        self.self.value = prune_linear_layer(self.self.value, index)
+        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
+        self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+    ):
+        h1, h2 = self.clone(hidden_states, 2)
+        self_outputs = self.self(
+            h1,
+            attention_mask,
+            head_mask,
+            encoder_hidden_states,
+            encoder_attention_mask,
+            output_attentions,
+        )
+        attention_output = self.output(self_outputs[0], h2)
+        outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        # assuming that we don't ouput the attentions (outputs = (attention_output,)), self_outputs=(context_layer,)
+        (cam1, cam2) = self.output.relprop(cam, **kwargs)
+        #print(cam1.sum(), cam2.sum(), (cam1 + cam2).sum())
+        cam1 = self.self.relprop(cam1, **kwargs)
+        #print(cam1.sum(), cam2.sum(), (cam1 + cam2).sum())
+
+        return self.clone.relprop((cam1, cam2), **kwargs)
+
+class BertSelfAttention(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+            raise ValueError(
+                "The hidden size (%d) is not a multiple of the number of attention "
+                "heads (%d)" % (config.hidden_size, config.num_attention_heads)
+            )
+
+        self.num_attention_heads = config.num_attention_heads
+        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        self.query = Linear(config.hidden_size, self.all_head_size)
+        self.key = Linear(config.hidden_size, self.all_head_size)
+        self.value = Linear(config.hidden_size, self.all_head_size)
+
+        self.dropout = Dropout(config.attention_probs_dropout_prob)
+
+        self.matmul1 = MatMul()
+        self.matmul2 = MatMul()
+        self.softmax = Softmax(dim=-1)
+        self.add = Add()
+        self.mul = Mul()
+        self.head_mask = None
+        self.attention_mask = None
+        self.clone = Clone()
+
+        self.attn_cam = None
+        self.attn = None
+        self.attn_gradients = None
+
+    def get_attn(self):
+        return self.attn
+
+    def save_attn(self, attn):
+        self.attn = attn
+
+    def save_attn_cam(self, cam):
+        self.attn_cam = cam
+
+    def get_attn_cam(self):
+        return self.attn_cam
+
+    def save_attn_gradients(self, attn_gradients):
+        self.attn_gradients = attn_gradients
+
+    def get_attn_gradients(self):
+        return self.attn_gradients
+
+    def transpose_for_scores(self, x):
+        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+        x = x.view(*new_x_shape)
+        return x.permute(0, 2, 1, 3)
+
+    def transpose_for_scores_relprop(self, x):
+        return x.permute(0, 2, 1, 3).flatten(2)
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=False,
+    ):
+        self.head_mask = head_mask
+        self.attention_mask = attention_mask
+
+        h1, h2, h3 = self.clone(hidden_states, 3)
+        mixed_query_layer = self.query(h1)
+
+        # If this is instantiated as a cross-attention module, the keys
+        # and values come from an encoder; the attention mask needs to be
+        # such that the encoder's padding tokens are not attended to.
+        if encoder_hidden_states is not None:
+            mixed_key_layer = self.key(encoder_hidden_states)
+            mixed_value_layer = self.value(encoder_hidden_states)
+            attention_mask = encoder_attention_mask
+        else:
+            mixed_key_layer = self.key(h2)
+            mixed_value_layer = self.value(h3)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer)
+        key_layer = self.transpose_for_scores(mixed_key_layer)
+        value_layer = self.transpose_for_scores(mixed_value_layer)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        attention_scores = self.matmul1([query_layer, key_layer.transpose(-1, -2)])
+        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
+            attention_scores = self.add([attention_scores, attention_mask])
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = self.softmax(attention_scores)
+
+        self.save_attn(attention_probs)
+        attention_probs.register_hook(self.save_attn_gradients)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = self.matmul2([attention_probs, value_layer])
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        context_layer = context_layer.view(*new_context_layer_shape)
+
+        outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        # Assume output_attentions == False
+        cam = self.transpose_for_scores(cam)
+
+        # [attention_probs, value_layer]
+        (cam1, cam2) = self.matmul2.relprop(cam, **kwargs)
+        cam1 /= 2
+        cam2 /= 2
+        if self.head_mask is not None:
+            # [attention_probs, head_mask]
+            (cam1, _)= self.mul.relprop(cam1, **kwargs)
+
+
+        self.save_attn_cam(cam1)
+
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+
+        cam1 = self.softmax.relprop(cam1, **kwargs)
+
+        if self.attention_mask is not None:
+            # [attention_scores, attention_mask]
+            (cam1, _) = self.add.relprop(cam1, **kwargs)
+
+        # [query_layer, key_layer.transpose(-1, -2)]
+        (cam1_1, cam1_2) = self.matmul1.relprop(cam1, **kwargs)
+        cam1_1 /= 2
+        cam1_2 /= 2
+
+        # query
+        cam1_1 = self.transpose_for_scores_relprop(cam1_1)
+        cam1_1 = self.query.relprop(cam1_1, **kwargs)
+
+        # key
+        cam1_2 = self.transpose_for_scores_relprop(cam1_2.transpose(-1, -2))
+        cam1_2 = self.key.relprop(cam1_2, **kwargs)
+
+        # value
+        cam2 = self.transpose_for_scores_relprop(cam2)
+        cam2 = self.value.relprop(cam2, **kwargs)
+
+        cam = self.clone.relprop((cam1_1, cam1_2, cam2), **kwargs)
+
+        return cam
+
+
+class BertSelfOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.hidden_size)
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.add = Add()
+
+    def forward(self, hidden_states, input_tensor):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        add = self.add([hidden_states, input_tensor])
+        hidden_states = self.LayerNorm(add)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+        # [hidden_states, input_tensor]
+        (cam1, cam2) = self.add.relprop(cam, **kwargs)
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+        cam1 = self.dense.relprop(cam1, **kwargs)
+
+        return (cam1, cam2)
+
+
+class BertIntermediate(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.hidden_size, config.intermediate_size)
+        if isinstance(config.hidden_act, str):
+            self.intermediate_act_fn = ACT2FN[config.hidden_act]()
+        else:
+            self.intermediate_act_fn = config.hidden_act
+
+    def forward(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.intermediate_act_fn(hidden_states)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        cam = self.intermediate_act_fn.relprop(cam, **kwargs)  # FIXME only ReLU
+        #print(cam.sum())
+        cam = self.dense.relprop(cam, **kwargs)
+        #print(cam.sum())
+        return cam
+
+
+class BertOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = Linear(config.intermediate_size, config.hidden_size)
+        self.LayerNorm = LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.add = Add()
+
+    def forward(self, hidden_states, input_tensor):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        add = self.add([hidden_states, input_tensor])
+        hidden_states = self.LayerNorm(add)
+        return hidden_states
+
+    def relprop(self, cam, **kwargs):
+        # print("in", cam.sum())
+        cam = self.LayerNorm.relprop(cam, **kwargs)
+        #print(cam.sum())
+        # [hidden_states, input_tensor]
+        (cam1, cam2)= self.add.relprop(cam, **kwargs)
+        # print("add", cam1.sum(), cam2.sum(), cam1.sum() + cam2.sum())
+        cam1 = self.dropout.relprop(cam1, **kwargs)
+        #print(cam1.sum())
+        cam1 = self.dense.relprop(cam1, **kwargs)
+        # print("dense", cam1.sum())
+
+        # print("out", cam1.sum() + cam2.sum(), cam1.sum(), cam2.sum())
+        return (cam1, cam2)
+
+
+class BertLayer(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.attention = BertAttention(config)
+        self.intermediate = BertIntermediate(config)
+        self.output = BertOutput(config)
+        self.clone = Clone()
+
+    def forward(
+            self,
+            hidden_states,
+            attention_mask=None,
+            head_mask=None,
+            output_attentions=False,
+    ):
+        self_attention_outputs = self.attention(
+            hidden_states,
+            attention_mask,
+            head_mask,
+            output_attentions=output_attentions,
+        )
+        attention_output = self_attention_outputs[0]
+        outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights
+
+        ao1, ao2 = self.clone(attention_output, 2)
+        intermediate_output = self.intermediate(ao1)
+        layer_output = self.output(intermediate_output, ao2)
+
+        outputs = (layer_output,) + outputs
+        return outputs
+
+    def relprop(self, cam, **kwargs):
+        (cam1, cam2) = self.output.relprop(cam, **kwargs)
+        # print("output", cam1.sum(), cam2.sum(), cam1.sum() + cam2.sum())
+        cam1 = self.intermediate.relprop(cam1, **kwargs)
+        # print("intermediate", cam1.sum())
+        cam = self.clone.relprop((cam1, cam2), **kwargs)
+        # print("clone", cam.sum())
+        cam = self.attention.relprop(cam, **kwargs)
+        # print("attention", cam.sum())
+        return cam
+
+
+class BertModel(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = BertEmbeddings(config)
+        self.encoder = BertEncoder(config)
+        self.pooler = BertPooler(config)
+
+        self.init_weights()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    def forward(
+            self,
+            input_ids=None,
+            attention_mask=None,
+            token_type_ids=None,
+            position_ids=None,
+            head_mask=None,
+            inputs_embeds=None,
+            encoder_hidden_states=None,
+            encoder_attention_mask=None,
+            output_attentions=None,
+            output_hidden_states=None,
+            return_dict=None,
+    ):
+        r"""
+        encoder_hidden_states  (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
+            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention
+            if the model is configured as a decoder.
+        encoder_attention_mask (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Mask to avoid performing attention on the padding token indices of the encoder input. This mask
+            is used in the cross-attention if the model is configured as a decoder.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if attention_mask is None:
+            attention_mask = torch.ones(input_shape, device=device)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape, device)
+
+        # If a 2D or 3D attention mask is provided for the cross-attention
+        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+        if self.config.is_decoder and encoder_hidden_states is not None:
+            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
+            if encoder_attention_mask is None:
+                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
+            encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+        else:
+            encoder_extended_attention_mask = None
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids, position_ids=position_ids, token_type_ids=token_type_ids, inputs_embeds=inputs_embeds
+        )
+
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=extended_attention_mask,
+            head_mask=head_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_extended_attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        sequence_output = encoder_outputs[0]
+        pooled_output = self.pooler(sequence_output)
+
+        if not return_dict:
+            return (sequence_output, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=sequence_output,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+    def relprop(self, cam, **kwargs):
+        cam = self.pooler.relprop(cam, **kwargs)
+        # print("111111111111",cam.sum())
+        cam = self.encoder.relprop(cam, **kwargs)
+        # print("222222222222222", cam.sum())
+        # print("conservation: ", cam.sum())
+        return cam
+
+
+if __name__ == '__main__':
+    class Config:
+      def __init__(self, hidden_size, num_attention_heads, attention_probs_dropout_prob):
+        self.hidden_size = hidden_size
+        self.num_attention_heads = num_attention_heads
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+
+    model = BertSelfAttention(Config(1024, 4, 0.1))
+    x = torch.rand(2, 20, 1024)
+    x.requires_grad_()
+
+    model.eval()
+
+    y = model.forward(x)
+
+    relprop = model.relprop(torch.rand(2, 20, 1024), (torch.rand(2, 20, 1024),))
+
+    print(relprop[1][0].shape)

BERT/BERT_explainability/modules/BERT/BertForSequenceClassification.py

@@ -0,0 +1,215 @@
+from turtle import forward
+
+from transformers import BertPreTrainedModel
+
+from BERT_explainability.modules.layers_ours import *
+from BERT_explainability.modules.BERT.BERT import BertModel
+from torch.nn import CrossEntropyLoss, MSELoss
+import torch.nn as nn
+from typing import List, Any
+import torch
+
+from BERT_rationale_benchmark.models.model_utils import PaddedSequence
+
+import sys
+sys.path.append("../../")
+from sequenceoutput.modeling_output import SequenceClassifierOutput
+
+class BertForSequenceClassification(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.bert = BertModel(config)
+        self.dropout = Dropout(config.hidden_dropout_prob)
+        self.classifier = Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+
+
+
+    def forward(
+            self,
+            input_ids=None,
+            attention_mask=None,
+            token_type_ids=None,
+            position_ids=None,
+            head_mask=None,
+            inputs_embeds=None,
+            labels=None,
+            output_attentions=None,
+            output_hidden_states=None,
+            return_dict=None,
+    ):
+        r"""
+        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`):
+            Labels for computing the sequence classification/regression loss.
+            Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
+            If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
+            If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(logits.view(-1), labels.view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+
+
+    def relprop(self, cam=None, **kwargs):
+        cam = self.classifier.relprop(cam, **kwargs)
+        cam = self.dropout.relprop(cam, **kwargs)
+        cam = self.bert.relprop(cam, **kwargs)
+        # print("conservation: ", cam.sum())
+        return cam
+
+
+# this is the actual classifier we will be using
+class BertClassifier(nn.Module):
+    """Thin wrapper around BertForSequenceClassification"""
+
+    def __init__(self,
+                 bert_dir: str,
+                 pad_token_id: int,
+                 cls_token_id: int,
+                 sep_token_id: int,
+                 num_labels: int,
+                 max_length: int = 512,
+                 use_half_precision=True):
+        super(BertClassifier, self).__init__()
+        bert = BertForSequenceClassification.from_pretrained(bert_dir, num_labels=num_labels)
+        if use_half_precision:
+            import apex
+            bert = bert.half()
+        self.bert = bert
+        self.pad_token_id = pad_token_id
+        self.cls_token_id = cls_token_id
+        self.sep_token_id = sep_token_id
+        self.max_length = max_length
+
+    def forward(self,
+                query: List[torch.tensor],
+                docids: List[Any],
+                document_batch: List[torch.tensor]):
+        assert len(query) == len(document_batch)
+        print(query)
+        # note about device management:
+        # since distributed training is enabled, the inputs to this module can be on *any* device (preferably cpu, since we wrap and unwrap the module)
+        # we want to keep these params on the input device (assuming CPU) for as long as possible for cheap memory access
+        target_device = next(self.parameters()).device
+        cls_token = torch.tensor([self.cls_token_id]).to(device=document_batch[0].device)
+        sep_token = torch.tensor([self.sep_token_id]).to(device=document_batch[0].device)
+        input_tensors = []
+        position_ids = []
+        for q, d in zip(query, document_batch):
+            if len(q) + len(d) + 2 > self.max_length:
+                d = d[:(self.max_length - len(q) - 2)]
+            input_tensors.append(torch.cat([cls_token, q, sep_token, d]))
+            position_ids.append(torch.tensor(list(range(0, len(q) + 1)) + list(range(0, len(d) + 1))))
+        bert_input = PaddedSequence.autopad(input_tensors, batch_first=True, padding_value=self.pad_token_id,
+                                            device=target_device)
+        positions = PaddedSequence.autopad(position_ids, batch_first=True, padding_value=0, device=target_device)
+        (classes,) = self.bert(bert_input.data,
+                               attention_mask=bert_input.mask(on=0.0, off=float('-inf'), device=target_device),
+                               position_ids=positions.data)
+        assert torch.all(classes == classes)  # for nans
+
+        print(input_tensors[0])
+        print(self.relprop()[0])
+
+        return classes
+
+    def relprop(self, cam=None, **kwargs):
+        return self.bert.relprop(cam, **kwargs)
+
+
+if __name__ == '__main__':
+    from transformers import BertTokenizer
+    import os
+
+    class Config:
+        def __init__(self, hidden_size, num_attention_heads, attention_probs_dropout_prob, num_labels,
+                     hidden_dropout_prob):
+            self.hidden_size = hidden_size
+            self.num_attention_heads = num_attention_heads
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.num_labels = num_labels
+            self.hidden_dropout_prob = hidden_dropout_prob
+
+
+    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
+    x = tokenizer.encode_plus("In this movie the acting is great. The movie is perfect! [sep]",
+                         add_special_tokens=True,
+                         max_length=512,
+                         return_token_type_ids=False,
+                         return_attention_mask=True,
+                         pad_to_max_length=True,
+                         return_tensors='pt',
+                         truncation=True)
+
+    print(x['input_ids'])
+
+    model = BertForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=2)
+    model_save_file = os.path.join('./BERT_explainability/output_bert/movies/classifier/', 'classifier.pt')
+    model.load_state_dict(torch.load(model_save_file))
+
+    # x = torch.randint(100, (2, 20))
+    # x = torch.tensor([[101, 2054, 2003, 1996, 15792, 1997, 2023, 3319, 1029, 102,
+    #                    101, 4079, 102, 101, 6732, 102, 101, 2643, 102, 101,
+    #                    2038, 102, 101, 1037, 102, 101, 2933, 102, 101, 2005,
+    #                    102, 101, 2032, 102, 101, 1010, 102, 101, 1037, 102,
+    #                    101, 3800, 102, 101, 2005, 102, 101, 2010, 102, 101,
+    #                    2166, 102, 101, 1010, 102, 101, 1998, 102, 101, 2010,
+    #                    102, 101, 4650, 102, 101, 1010, 102, 101, 2002, 102,
+    #                    101, 2074, 102, 101, 2515, 102, 101, 1050, 102, 101,
+    #                    1005, 102, 101, 1056, 102, 101, 2113, 102, 101, 2054,
+    #                    102, 101, 1012, 102]])
+    # x.requires_grad_()
+
+    model.eval()
+
+    y = model(x['input_ids'], x['attention_mask'])
+    print(y)
+
+    cam, _ = model.relprop()
+
+    #print(cam.shape)
+
+    cam = cam.sum(-1)
+    #print(cam)

BERT/BERT_explainability/modules/BERT/ExplanationGenerator.py

@@ -0,0 +1,156 @@
+import argparse
+import numpy as np
+import torch
+import glob
+
+# compute rollout between attention layers
+def compute_rollout_attention(all_layer_matrices, start_layer=0):
+    # adding residual consideration- code adapted from https://github.com/samiraabnar/attention_flow
+    num_tokens = all_layer_matrices[0].shape[1]
+    batch_size = all_layer_matrices[0].shape[0]
+    eye = torch.eye(num_tokens).expand(batch_size, num_tokens, num_tokens).to(all_layer_matrices[0].device)
+    all_layer_matrices = [all_layer_matrices[i] + eye for i in range(len(all_layer_matrices))]
+    matrices_aug = [all_layer_matrices[i] / all_layer_matrices[i].sum(dim=-1, keepdim=True)
+                          for i in range(len(all_layer_matrices))]
+    joint_attention = matrices_aug[start_layer]
+    for i in range(start_layer+1, len(matrices_aug)):
+        joint_attention = matrices_aug[i].bmm(joint_attention)
+    return joint_attention
+
+class Generator:
+    def __init__(self, model):
+        self.model = model
+        self.model.eval()
+
+    def forward(self, input_ids, attention_mask):
+        return self.model(input_ids, attention_mask)
+
+    def generate_LRP(self, input_ids, attention_mask,
+                     index=None, start_layer=11):
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        kwargs = {"alpha": 1}
+
+        if index == None:
+            index = np.argmax(output.cpu().data.numpy(), axis=-1)
+
+        one_hot = np.zeros((1, output.size()[-1]), dtype=np.float32)
+        one_hot[0, index] = 1
+        one_hot_vector = one_hot
+        one_hot = torch.from_numpy(one_hot).requires_grad_(True)
+        one_hot = torch.sum(one_hot.cuda() * output)
+
+        self.model.zero_grad()
+        one_hot.backward(retain_graph=True)
+
+        self.model.relprop(torch.tensor(one_hot_vector).to(input_ids.device), **kwargs)
+
+        cams = []
+        blocks = self.model.bert.encoder.layer
+        for blk in blocks:
+            grad = blk.attention.self.get_attn_gradients()
+            cam = blk.attention.self.get_attn_cam()
+            cam = cam[0].reshape(-1, cam.shape[-1], cam.shape[-1])
+            grad = grad[0].reshape(-1, grad.shape[-1], grad.shape[-1])
+            cam = grad * cam
+            cam = cam.clamp(min=0).mean(dim=0)
+            cams.append(cam.unsqueeze(0))
+        rollout = compute_rollout_attention(cams, start_layer=start_layer)
+        rollout[:, 0, 0] = rollout[:, 0].min()
+        return rollout[:, 0]
+
+
+    def generate_LRP_last_layer(self, input_ids, attention_mask,
+                     index=None):
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        kwargs = {"alpha": 1}
+        if index == None:
+            index = np.argmax(output.cpu().data.numpy(), axis=-1)
+
+        one_hot = np.zeros((1, output.size()[-1]), dtype=np.float32)
+        one_hot[0, index] = 1
+        one_hot_vector = one_hot
+        one_hot = torch.from_numpy(one_hot).requires_grad_(True)
+        one_hot = torch.sum(one_hot.cuda() * output)
+
+        self.model.zero_grad()
+        one_hot.backward(retain_graph=True)
+
+        self.model.relprop(torch.tensor(one_hot_vector).to(input_ids.device), **kwargs)
+
+        cam = self.model.bert.encoder.layer[-1].attention.self.get_attn_cam()[0]
+        cam = cam.clamp(min=0).mean(dim=0).unsqueeze(0)
+        cam[:, 0, 0] = 0
+        return cam[:, 0]
+
+    def generate_full_lrp(self, input_ids, attention_mask,
+                     index=None):
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        kwargs = {"alpha": 1}
+
+        if index == None:
+            index = np.argmax(output.cpu().data.numpy(), axis=-1)
+
+        one_hot = np.zeros((1, output.size()[-1]), dtype=np.float32)
+        one_hot[0, index] = 1
+        one_hot_vector = one_hot
+        one_hot = torch.from_numpy(one_hot).requires_grad_(True)
+        one_hot = torch.sum(one_hot.cuda() * output)
+
+        self.model.zero_grad()
+        one_hot.backward(retain_graph=True)
+
+        cam = self.model.relprop(torch.tensor(one_hot_vector).to(input_ids.device), **kwargs)
+        cam = cam.sum(dim=2)
+        cam[:, 0] = 0
+        return cam
+
+    def generate_attn_last_layer(self, input_ids, attention_mask,
+                     index=None):
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        cam = self.model.bert.encoder.layer[-1].attention.self.get_attn()[0]
+        cam = cam.mean(dim=0).unsqueeze(0)
+        cam[:, 0, 0] = 0
+        return cam[:, 0]
+
+    def generate_rollout(self, input_ids, attention_mask, start_layer=0, index=None):
+        self.model.zero_grad()
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        blocks = self.model.bert.encoder.layer
+        all_layer_attentions = []
+        for blk in blocks:
+            attn_heads = blk.attention.self.get_attn()
+            avg_heads = (attn_heads.sum(dim=1) / attn_heads.shape[1]).detach()
+            all_layer_attentions.append(avg_heads)
+        rollout = compute_rollout_attention(all_layer_attentions, start_layer=start_layer)
+        rollout[:, 0, 0] = 0
+        return rollout[:, 0]
+
+    def generate_attn_gradcam(self, input_ids, attention_mask, index=None):
+        output = self.model(input_ids=input_ids, attention_mask=attention_mask)[0]
+        kwargs = {"alpha": 1}
+
+        if index == None:
+            index = np.argmax(output.cpu().data.numpy(), axis=-1)
+
+        one_hot = np.zeros((1, output.size()[-1]), dtype=np.float32)
+        one_hot[0, index] = 1
+        one_hot_vector = one_hot
+        one_hot = torch.from_numpy(one_hot).requires_grad_(True)
+        one_hot = torch.sum(one_hot.cuda() * output)
+
+        self.model.zero_grad()
+        one_hot.backward(retain_graph=True)
+
+        self.model.relprop(torch.tensor(one_hot_vector).to(input_ids.device), **kwargs)
+
+        cam = self.model.bert.encoder.layer[-1].attention.self.get_attn()
+        grad = self.model.bert.encoder.layer[-1].attention.self.get_attn_gradients()
+
+        cam = cam[0].reshape(-1, cam.shape[-1], cam.shape[-1])
+        grad = grad[0].reshape(-1, grad.shape[-1], grad.shape[-1])
+        grad = grad.mean(dim=[1, 2], keepdim=True)
+        cam = (cam * grad).mean(0).clamp(min=0).unsqueeze(0)
+        cam = (cam - cam.min()) / (cam.max() - cam.min())
+        cam[:, 0, 0] = 0
+        return cam[:, 0]
+

BERT/BERT_explainability/modules/layers_lrp.py

@@ -0,0 +1,268 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+__all__ = ['forward_hook', 'Clone', 'Add', 'Cat', 'ReLU', 'GELU', 'Dropout', 'BatchNorm2d', 'Linear', 'MaxPool2d',
+           'AdaptiveAvgPool2d', 'AvgPool2d', 'Conv2d', 'Sequential', 'safe_divide', 'einsum', 'Softmax', 'IndexSelect',
+           'LayerNorm', 'AddEye', 'Tanh', 'MatMul', 'Mul']
+
+
+def safe_divide(a, b):
+    den = b.clamp(min=1e-9) + b.clamp(max=1e-9)
+    den = den + den.eq(0).type(den.type()) * 1e-9
+    return a / den * b.ne(0).type(b.type())
+
+
+def forward_hook(self, input, output):
+    if type(input[0]) in (list, tuple):
+        self.X = []
+        for i in input[0]:
+            x = i.detach()
+            x.requires_grad = True
+            self.X.append(x)
+    else:
+        self.X = input[0].detach()
+        self.X.requires_grad = True
+
+    self.Y = output
+
+
+def backward_hook(self, grad_input, grad_output):
+    self.grad_input = grad_input
+    self.grad_output = grad_output
+
+
+class RelProp(nn.Module):
+    def __init__(self):
+        super(RelProp, self).__init__()
+        # if not self.training:
+        self.register_forward_hook(forward_hook)
+
+    def gradprop(self, Z, X, S):
+        C = torch.autograd.grad(Z, X, S, retain_graph=True)
+        return C
+
+    def relprop(self, R, alpha):
+        return R
+
+
+class RelPropSimple(RelProp):
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        if torch.is_tensor(self.X) == False:
+            outputs = []
+            outputs.append(self.X[0] * C[0])
+            outputs.append(self.X[1] * C[1])
+        else:
+            outputs = self.X * (C[0])
+        return outputs
+
+class AddEye(RelPropSimple):
+    # input of shape B, C, seq_len, seq_len
+    def forward(self, input):
+        return input + torch.eye(input.shape[2]).expand_as(input).to(input.device)
+
+class ReLU(nn.ReLU, RelProp):
+    pass
+
+class Tanh(nn.Tanh, RelProp):
+    pass
+
+class GELU(nn.GELU, RelProp):
+    pass
+
+class Softmax(nn.Softmax, RelProp):
+    pass
+
+class LayerNorm(nn.LayerNorm, RelProp):
+    pass
+
+class Dropout(nn.Dropout, RelProp):
+    pass
+
+
+class MaxPool2d(nn.MaxPool2d, RelPropSimple):
+    pass
+
+class LayerNorm(nn.LayerNorm, RelProp):
+    pass
+
+class AdaptiveAvgPool2d(nn.AdaptiveAvgPool2d, RelPropSimple):
+    pass
+
+class MatMul(RelPropSimple):
+    def forward(self, inputs):
+        return torch.matmul(*inputs)
+
+class Mul(RelPropSimple):
+    def forward(self, inputs):
+        return torch.mul(*inputs)
+
+class AvgPool2d(nn.AvgPool2d, RelPropSimple):
+    pass
+
+
+class Add(RelPropSimple):
+    def forward(self, inputs):
+        return torch.add(*inputs)
+
+class einsum(RelPropSimple):
+    def __init__(self, equation):
+        super().__init__()
+        self.equation = equation
+    def forward(self, *operands):
+        return torch.einsum(self.equation, *operands)
+
+class IndexSelect(RelProp):
+    def forward(self, inputs, dim, indices):
+        self.__setattr__('dim', dim)
+        self.__setattr__('indices', indices)
+
+        return torch.index_select(inputs, dim, indices)
+
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X, self.dim, self.indices)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        if torch.is_tensor(self.X) == False:
+            outputs = []
+            outputs.append(self.X[0] * C[0])
+            outputs.append(self.X[1] * C[1])
+        else:
+            outputs = self.X * (C[0])
+        return outputs
+
+
+
+class Clone(RelProp):
+    def forward(self, input, num):
+        self.__setattr__('num', num)
+        outputs = []
+        for _ in range(num):
+            outputs.append(input)
+
+        return outputs
+
+    def relprop(self, R, alpha):
+        Z = []
+        for _ in range(self.num):
+            Z.append(self.X)
+        S = [safe_divide(r, z) for r, z in zip(R, Z)]
+        C = self.gradprop(Z, self.X, S)[0]
+
+        R = self.X * C
+
+        return R
+
+class Cat(RelProp):
+    def forward(self, inputs, dim):
+        self.__setattr__('dim', dim)
+        return torch.cat(inputs, dim)
+
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X, self.dim)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        outputs = []
+        for x, c in zip(self.X, C):
+            outputs.append(x * c)
+
+        return outputs
+
+class Sequential(nn.Sequential):
+    def relprop(self, R, alpha):
+        for m in reversed(self._modules.values()):
+            R = m.relprop(R, alpha)
+        return R
+
+class BatchNorm2d(nn.BatchNorm2d, RelProp):
+    def relprop(self, R, alpha):
+        X = self.X
+        beta = 1 - alpha
+        weight = self.weight.unsqueeze(0).unsqueeze(2).unsqueeze(3) / (
+            (self.running_var.unsqueeze(0).unsqueeze(2).unsqueeze(3).pow(2) + self.eps).pow(0.5))
+        Z = X * weight + 1e-9
+        S = R / Z
+        Ca = S * weight
+        R = self.X * (Ca)
+        return R
+
+
+class Linear(nn.Linear, RelProp):
+    def relprop(self, R, alpha):
+        beta = alpha - 1
+        pw = torch.clamp(self.weight, min=0)
+        nw = torch.clamp(self.weight, max=0)
+        px = torch.clamp(self.X, min=0)
+        nx = torch.clamp(self.X, max=0)
+
+        def f(w1, w2, x1, x2):
+            Z1 = F.linear(x1, w1)
+            Z2 = F.linear(x2, w2)
+            S1 = safe_divide(R, Z1)
+            S2 = safe_divide(R, Z2)
+            C1 = x1 * torch.autograd.grad(Z1, x1, S1)[0]
+            C2 = x2 * torch.autograd.grad(Z2, x2, S2)[0]
+
+            return C1 + C2
+
+        activator_relevances = f(pw, nw, px, nx)
+        inhibitor_relevances = f(nw, pw, px, nx)
+
+        R = alpha * activator_relevances - beta * inhibitor_relevances
+
+        return R
+
+class Conv2d(nn.Conv2d, RelProp):
+    def gradprop2(self, DY, weight):
+        Z = self.forward(self.X)
+
+        output_padding = self.X.size()[2] - (
+                (Z.size()[2] - 1) * self.stride[0] - 2 * self.padding[0] + self.kernel_size[0])
+
+        return F.conv_transpose2d(DY, weight, stride=self.stride, padding=self.padding, output_padding=output_padding)
+
+    def relprop(self, R, alpha):
+        if self.X.shape[1] == 3:
+            pw = torch.clamp(self.weight, min=0)
+            nw = torch.clamp(self.weight, max=0)
+            X = self.X
+            L = self.X * 0 + \
+                torch.min(torch.min(torch.min(self.X, dim=1, keepdim=True)[0], dim=2, keepdim=True)[0], dim=3,
+                          keepdim=True)[0]
+            H = self.X * 0 + \
+                torch.max(torch.max(torch.max(self.X, dim=1, keepdim=True)[0], dim=2, keepdim=True)[0], dim=3,
+                          keepdim=True)[0]
+            Za = torch.conv2d(X, self.weight, bias=None, stride=self.stride, padding=self.padding) - \
+                 torch.conv2d(L, pw, bias=None, stride=self.stride, padding=self.padding) - \
+                 torch.conv2d(H, nw, bias=None, stride=self.stride, padding=self.padding) + 1e-9
+
+            S = R / Za
+            C = X * self.gradprop2(S, self.weight) - L * self.gradprop2(S, pw) - H * self.gradprop2(S, nw)
+            R = C
+        else:
+            beta = alpha - 1
+            pw = torch.clamp(self.weight, min=0)
+            nw = torch.clamp(self.weight, max=0)
+            px = torch.clamp(self.X, min=0)
+            nx = torch.clamp(self.X, max=0)
+
+            def f(w1, w2, x1, x2):
+                Z1 = F.conv2d(x1, w1, bias=None, stride=self.stride, padding=self.padding)
+                Z2 = F.conv2d(x2, w2, bias=None, stride=self.stride, padding=self.padding)
+                S1 = safe_divide(R, Z1)
+                S2 = safe_divide(R, Z2)
+                C1 = x1 * self.gradprop(Z1, x1, S1)[0]
+                C2 = x2 * self.gradprop(Z2, x2, S2)[0]
+                return C1 + C2
+
+            activator_relevances = f(pw, nw, px, nx)
+            inhibitor_relevances = f(nw, pw, px, nx)
+
+            R = alpha * activator_relevances - beta * inhibitor_relevances
+        return R

BERT/BERT_explainability/modules/layers_ours.py

@@ -0,0 +1,292 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+__all__ = ['forward_hook', 'Clone', 'Add', 'Cat', 'ReLU', 'GELU', 'Dropout', 'BatchNorm2d', 'Linear', 'MaxPool2d',
+           'AdaptiveAvgPool2d', 'AvgPool2d', 'Conv2d', 'Sequential', 'safe_divide', 'einsum', 'Softmax', 'IndexSelect',
+           'LayerNorm', 'AddEye', 'Tanh', 'MatMul', 'Mul']
+
+
+def safe_divide(a, b):
+    den = b.clamp(min=1e-9) + b.clamp(max=1e-9)
+    den = den + den.eq(0).type(den.type()) * 1e-9
+    return a / den * b.ne(0).type(b.type())
+
+
+def forward_hook(self, input, output):
+    if type(input[0]) in (list, tuple):
+        self.X = []
+        for i in input[0]:
+            x = i.detach()
+            x.requires_grad = True
+            self.X.append(x)
+    else:
+        self.X = input[0].detach()
+        self.X.requires_grad = True
+
+    self.Y = output
+
+
+def backward_hook(self, grad_input, grad_output):
+    self.grad_input = grad_input
+    self.grad_output = grad_output
+
+
+class RelProp(nn.Module):
+    def __init__(self):
+        super(RelProp, self).__init__()
+        # if not self.training:
+        self.register_forward_hook(forward_hook)
+
+    def gradprop(self, Z, X, S):
+        C = torch.autograd.grad(Z, X, S, retain_graph=True)
+        return C
+
+    def relprop(self, R, alpha):
+        return R
+
+
+class RelPropSimple(RelProp):
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        if torch.is_tensor(self.X) == False:
+            outputs = []
+            outputs.append(self.X[0] * C[0])
+            outputs.append(self.X[1] * C[1])
+        else:
+            outputs = self.X * (C[0])
+        return outputs
+
+class AddEye(RelPropSimple):
+    # input of shape B, C, seq_len, seq_len
+    def forward(self, input):
+        return input + torch.eye(input.shape[2]).expand_as(input).to(input.device)
+
+class ReLU(nn.ReLU, RelProp):
+    pass
+
+class GELU(nn.GELU, RelProp):
+    pass
+
+class Softmax(nn.Softmax, RelProp):
+    pass
+
+class Mul(RelPropSimple):
+    def forward(self, inputs):
+        return torch.mul(*inputs)
+
+class Tanh(nn.Tanh, RelProp):
+    pass
+class LayerNorm(nn.LayerNorm, RelProp):
+    pass
+
+class Dropout(nn.Dropout, RelProp):
+    pass
+
+class MatMul(RelPropSimple):
+    def forward(self, inputs):
+        return torch.matmul(*inputs)
+
+class MaxPool2d(nn.MaxPool2d, RelPropSimple):
+    pass
+
+class LayerNorm(nn.LayerNorm, RelProp):
+    pass
+
+class AdaptiveAvgPool2d(nn.AdaptiveAvgPool2d, RelPropSimple):
+    pass
+
+
+class AvgPool2d(nn.AvgPool2d, RelPropSimple):
+    pass
+
+
+class Add(RelPropSimple):
+    def forward(self, inputs):
+        return torch.add(*inputs)
+
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        a = self.X[0] * C[0]
+        b = self.X[1] * C[1]
+
+        a_sum = a.sum()
+        b_sum = b.sum()
+
+        a_fact = safe_divide(a_sum.abs(), a_sum.abs() + b_sum.abs()) * R.sum()
+        b_fact = safe_divide(b_sum.abs(), a_sum.abs() + b_sum.abs()) * R.sum()
+
+        a = a * safe_divide(a_fact, a.sum())
+        b = b * safe_divide(b_fact, b.sum())
+
+        outputs = [a, b]
+
+        return outputs
+
+class einsum(RelPropSimple):
+    def __init__(self, equation):
+        super().__init__()
+        self.equation = equation
+    def forward(self, *operands):
+        return torch.einsum(self.equation, *operands)
+
+class IndexSelect(RelProp):
+    def forward(self, inputs, dim, indices):
+        self.__setattr__('dim', dim)
+        self.__setattr__('indices', indices)
+
+        return torch.index_select(inputs, dim, indices)
+
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X, self.dim, self.indices)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        if torch.is_tensor(self.X) == False:
+            outputs = []
+            outputs.append(self.X[0] * C[0])
+            outputs.append(self.X[1] * C[1])
+        else:
+            outputs = self.X * (C[0])
+        return outputs
+
+
+
+class Clone(RelProp):
+    def forward(self, input, num):
+        self.__setattr__('num', num)
+        outputs = []
+        for _ in range(num):
+            outputs.append(input)
+
+        return outputs
+
+    def relprop(self, R, alpha):
+        Z = []
+        for _ in range(self.num):
+            Z.append(self.X)
+        S = [safe_divide(r, z) for r, z in zip(R, Z)]
+        C = self.gradprop(Z, self.X, S)[0]
+
+        R = self.X * C
+
+        return R
+
+
+class Cat(RelProp):
+    def forward(self, inputs, dim):
+        self.__setattr__('dim', dim)
+        return torch.cat(inputs, dim)
+
+    def relprop(self, R, alpha):
+        Z = self.forward(self.X, self.dim)
+        S = safe_divide(R, Z)
+        C = self.gradprop(Z, self.X, S)
+
+        outputs = []
+        for x, c in zip(self.X, C):
+            outputs.append(x * c)
+
+        return outputs
+
+
+class Sequential(nn.Sequential):
+    def relprop(self, R, alpha):
+        for m in reversed(self._modules.values()):
+            R = m.relprop(R, alpha)
+        return R
+
+
+class BatchNorm2d(nn.BatchNorm2d, RelProp):
+    def relprop(self, R, alpha):
+        X = self.X
+        beta = 1 - alpha
+        weight = self.weight.unsqueeze(0).unsqueeze(2).unsqueeze(3) / (
+            (self.running_var.unsqueeze(0).unsqueeze(2).unsqueeze(3).pow(2) + self.eps).pow(0.5))
+        Z = X * weight + 1e-9
+        S = R / Z
+        Ca = S * weight
+        R = self.X * (Ca)
+        return R
+
+
+class Linear(nn.Linear, RelProp):
+    def relprop(self, R, alpha):
+        beta = alpha - 1
+        pw = torch.clamp(self.weight, min=0)
+        nw = torch.clamp(self.weight, max=0)
+        px = torch.clamp(self.X, min=0)
+        nx = torch.clamp(self.X, max=0)
+
+        def f(w1, w2, x1, x2):
+            Z1 = F.linear(x1, w1)
+            Z2 = F.linear(x2, w2)
+            S1 = safe_divide(R, Z1 + Z2)
+            S2 = safe_divide(R, Z1 + Z2)
+            C1 = x1 * self.gradprop(Z1, x1, S1)[0]
+            C2 = x2 * self.gradprop(Z2, x2, S2)[0]
+
+            return C1 + C2
+
+        activator_relevances = f(pw, nw, px, nx)
+        inhibitor_relevances = f(nw, pw, px, nx)
+
+        R = alpha * activator_relevances - beta * inhibitor_relevances
+
+        return R
+
+
+class Conv2d(nn.Conv2d, RelProp):
+    def gradprop2(self, DY, weight):
+        Z = self.forward(self.X)
+
+        output_padding = self.X.size()[2] - (
+                (Z.size()[2] - 1) * self.stride[0] - 2 * self.padding[0] + self.kernel_size[0])
+
+        return F.conv_transpose2d(DY, weight, stride=self.stride, padding=self.padding, output_padding=output_padding)
+
+    def relprop(self, R, alpha):
+        if self.X.shape[1] == 3:
+            pw = torch.clamp(self.weight, min=0)
+            nw = torch.clamp(self.weight, max=0)
+            X = self.X
+            L = self.X * 0 + \
+                torch.min(torch.min(torch.min(self.X, dim=1, keepdim=True)[0], dim=2, keepdim=True)[0], dim=3,
+                          keepdim=True)[0]
+            H = self.X * 0 + \
+                torch.max(torch.max(torch.max(self.X, dim=1, keepdim=True)[0], dim=2, keepdim=True)[0], dim=3,
+                          keepdim=True)[0]
+            Za = torch.conv2d(X, self.weight, bias=None, stride=self.stride, padding=self.padding) - \
+                 torch.conv2d(L, pw, bias=None, stride=self.stride, padding=self.padding) - \
+                 torch.conv2d(H, nw, bias=None, stride=self.stride, padding=self.padding) + 1e-9
+
+            S = R / Za
+            C = X * self.gradprop2(S, self.weight) - L * self.gradprop2(S, pw) - H * self.gradprop2(S, nw)
+            R = C
+        else:
+            beta = alpha - 1
+            pw = torch.clamp(self.weight, min=0)
+            nw = torch.clamp(self.weight, max=0)
+            px = torch.clamp(self.X, min=0)
+            nx = torch.clamp(self.X, max=0)
+
+            def f(w1, w2, x1, x2):
+                Z1 = F.conv2d(x1, w1, bias=None, stride=self.stride, padding=self.padding)
+                Z2 = F.conv2d(x2, w2, bias=None, stride=self.stride, padding=self.padding)
+                S1 = safe_divide(R, Z1)
+                S2 = safe_divide(R, Z2)
+                C1 = x1 * self.gradprop(Z1, x1, S1)[0]
+                C2 = x2 * self.gradprop(Z2, x2, S2)[0]
+                return C1 + C2
+
+            activator_relevances = f(pw, nw, px, nx)
+            inhibitor_relevances = f(nw, pw, px, nx)
+
+            R = alpha * activator_relevances - beta * inhibitor_relevances
+        return R

BERT/BERT_params/boolq.json

@@ -0,0 +1,26 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.05
+    },
+    "evidence_identifier": {
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "False", "True" ],
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "everything"
+    }
+}

BERT/BERT_params/boolq_baas.json

@@ -0,0 +1,26 @@
+{
+    "start_server": 0,
+    "bert_dir": "model_components/uncased_L-12_H-768_A-12/",
+    "max_length": 512,
+    "pooling_strategy": "CLS_TOKEN",
+    "evidence_identifier": {
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "False", "True" ],
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "everything"
+    }
+}
+
+

BERT/BERT_params/boolq_bert.json

@@ -0,0 +1,32 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 1,
+    "use_evidence_token_identifier": 0,
+    "evidence_identifier": {
+        "batch_size": 10,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 50,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "random",
+        "sampling_ratio": 1,
+        "use_half_precision": 0
+    },
+    "evidence_classifier": {
+        "classes": [
+            "False",
+            "True"
+        ],
+        "batch_size": 10,
+        "warmup_steps": 50,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/boolq_soft.json

@@ -0,0 +1,21 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.2
+    },
+    "classifier": {
+        "classes": [ "False", "True" ],
+        "has_query": 1,
+        "hidden_size": 32,
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 16,
+        "epochs": 50,
+        "attention_epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "dropout": 0.2,
+        "k_fraction": 0.07,
+        "threshold": 0.1
+    }
+}

BERT/BERT_params/cose_bert.json

@@ -0,0 +1,30 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 0,
+    "use_evidence_token_identifier": 1,
+    "evidence_token_identifier": {
+        "batch_size": 32,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 0.5,
+        "sampling_method": "everything",
+        "use_half_precision": 0,
+        "cose_data_hack": 1
+    },
+    "evidence_classifier": {
+        "classes": [ "false", "true"],
+        "batch_size": 32,
+        "warmup_steps": 10,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 0.5,
+        "sampling_method": "everything",
+        "use_half_precision": 0,
+        "cose_data_hack": 1
+    }
+}

BERT/BERT_params/cose_multiclass.json

@@ -0,0 +1,35 @@
+{
+  "max_length": 512,
+  "bert_vocab": "bert-base-uncased",
+  "bert_dir": "bert-base-uncased",
+  "use_evidence_sentence_identifier": 1,
+  "use_evidence_token_identifier": 0,
+  "evidence_identifier": {
+    "batch_size": 32,
+    "epochs": 10,
+    "patience": 10,
+    "warmup_steps": 50,
+    "lr": 1e-05,
+    "max_grad_norm": 1,
+    "sampling_method": "random",
+    "sampling_ratio": 1,
+    "use_half_precision": 0
+  },
+  "evidence_classifier": {
+    "classes": [
+      "A",
+      "B",
+      "C",
+      "D",
+      "E"
+    ],
+    "batch_size": 10,
+    "warmup_steps": 50,
+    "epochs": 10,
+    "patience": 10,
+    "lr": 1e-05,
+    "max_grad_norm": 1,
+    "sampling_method": "everything",
+    "use_half_precision": 0
+  }
+}

BERT/BERT_params/esnli_bert.json

@@ -0,0 +1,28 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 0,
+    "use_evidence_token_identifier": 1,
+    "evidence_token_identifier": {
+        "batch_size": 32,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    },
+    "evidence_classifier": {
+        "classes": [ "contradiction", "neutral", "entailment" ],
+        "batch_size": 32,
+        "warmup_steps": 10,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/evidence_inference.json

@@ -0,0 +1,26 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/PubMed-w2v.bin",
+        "dropout": 0.05
+    },
+    "evidence_identifier": {
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "significantly decreased", "no significant difference", "significantly increased" ],
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "everything"
+    }
+}

BERT/BERT_params/evidence_inference_bert.json

@@ -0,0 +1,33 @@
+{
+    "max_length": 512,
+    "bert_vocab": "allenai/scibert_scivocab_uncased",
+    "bert_dir": "allenai/scibert_scivocab_uncased",
+    "use_evidence_sentence_identifier": 1,
+    "use_evidence_token_identifier": 0,
+    "evidence_identifier": {
+        "batch_size": 10,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "random",
+        "use_half_precision": 0,
+        "sampling_ratio": 1
+    },
+    "evidence_classifier": {
+        "classes": [
+            "significantly decreased",
+            "no significant difference",
+            "significantly increased"
+        ],
+        "batch_size": 10,
+        "warmup_steps": 10,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/evidence_inference_soft.json

@@ -0,0 +1,22 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/PubMed-w2v.bin",
+        "dropout": 0.2
+    },
+    "classifier": {
+        "classes": [ "significantly decreased", "no significant difference", "significantly increased" ],
+        "use_token_selection": 1,
+        "has_query": 1,
+        "hidden_size": 32,
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 16,
+        "epochs": 50,
+        "attention_epochs": 0,
+        "patience": 10,
+        "lr": 1e-3,
+        "dropout": 0.2,
+        "k_fraction": 0.013,
+        "threshold": 0.1
+    }
+}

BERT/BERT_params/fever.json

@@ -0,0 +1,26 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.05
+    },
+    "evidence_identifier": {
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "SUPPORTS", "REFUTES" ],
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-5,
+        "sampling_method": "everything"
+    }
+}

BERT/BERT_params/fever_baas.json

@@ -0,0 +1,25 @@
+{
+    "start_server": 0,
+    "bert_dir": "model_components/uncased_L-12_H-768_A-12/",
+    "max_length": 512,
+    "pooling_strategy": "CLS_TOKEN",
+    "evidence_identifier": {
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "SUPPORTS", "REFUTES" ],
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "everything"
+    }
+}
+

BERT/BERT_params/fever_bert.json

@@ -0,0 +1,32 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 1,
+    "use_evidence_token_identifier": 0,
+    "evidence_identifier": {
+        "batch_size": 16,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1.0,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0,
+        "use_half_precision": 0
+    },
+    "evidence_classifier": {
+        "classes": [
+            "SUPPORTS",
+            "REFUTES"
+        ],
+        "batch_size": 10,
+        "warmup_steps": 10,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1.0,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/fever_soft.json

@@ -0,0 +1,21 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.2
+    },
+    "classifier": {
+        "classes": [ "SUPPORTS", "REFUTES" ],
+        "has_query": 1,
+        "hidden_size": 32,
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 128,
+        "epochs": 50,
+        "attention_epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "dropout": 0.2,
+        "k_fraction": 0.07,
+        "threshold": 0.1
+    }
+}

BERT/BERT_params/movies.json

@@ -0,0 +1,26 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.05
+    },
+    "evidence_identifier": {
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-4,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "NEG", "POS" ],
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "everything"
+    }
+}

BERT/BERT_params/movies_baas.json

@@ -0,0 +1,26 @@
+{
+    "start_server": 0,
+    "bert_dir": "model_components/uncased_L-12_H-768_A-12/",
+    "max_length": 512,
+    "pooling_strategy": "CLS_TOKEN",
+    "evidence_identifier": {
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "NEG", "POS" ],
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "everything"
+    }
+}
+
+

BERT/BERT_params/movies_bert.json

@@ -0,0 +1,32 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 1,
+    "use_evidence_token_identifier": 0,
+    "evidence_identifier": {
+        "batch_size": 16,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 50,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "random",
+        "sampling_ratio": 1,
+        "use_half_precision": 0
+    },
+    "evidence_classifier": {
+        "classes": [
+            "NEG",
+            "POS"
+        ],
+        "batch_size": 10,
+        "warmup_steps": 50,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/movies_soft.json

@@ -0,0 +1,21 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.2
+    },
+    "classifier": {
+        "classes": [ "NEG", "POS" ],
+        "has_query": 0,
+        "hidden_size": 32,
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 16,
+        "epochs": 50,
+        "attention_epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "dropout": 0.2,
+        "k_fraction": 0.07,
+        "threshold": 0.1
+    }
+}

BERT/BERT_params/multirc.json

@@ -0,0 +1,26 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.05
+    },
+    "evidence_identifier": {
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "False", "True" ],
+        "mlp_size": 128,
+        "dropout": 0.05,
+        "batch_size": 768,
+        "epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "sampling_method": "everything"
+    }
+}

BERT/BERT_params/multirc_baas.json

@@ -0,0 +1,26 @@
+{
+    "start_server": 0,
+    "bert_dir": "model_components/uncased_L-12_H-768_A-12/",
+    "max_length": 512,
+    "pooling_strategy": "CLS_TOKEN",
+    "evidence_identifier": {
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "random",
+        "sampling_ratio": 1.0
+    },
+    "evidence_classifier": {
+        "classes": [ "False", "True" ],
+        "batch_size": 64,
+        "epochs": 3,
+        "patience": 10,
+        "lr": 1e-3,
+        "max_grad_norm": 1.0,
+        "sampling_method": "everything"
+    }
+}
+
+

BERT/BERT_params/multirc_bert.json

@@ -0,0 +1,32 @@
+{
+    "max_length": 512,
+    "bert_vocab": "bert-base-uncased",
+    "bert_dir": "bert-base-uncased",
+    "use_evidence_sentence_identifier": 1,
+    "use_evidence_token_identifier": 0,
+    "evidence_identifier": {
+        "batch_size": 32,
+        "epochs": 10,
+        "patience": 10,
+        "warmup_steps": 50,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "random",
+        "sampling_ratio": 1,
+        "use_half_precision": 0
+    },
+    "evidence_classifier": {
+        "classes": [
+            "False",
+            "True"
+        ],
+        "batch_size": 32,
+        "warmup_steps": 50,
+        "epochs": 10,
+        "patience": 10,
+        "lr": 1e-05,
+        "max_grad_norm": 1,
+        "sampling_method": "everything",
+        "use_half_precision": 0
+    }
+}

BERT/BERT_params/multirc_soft.json

@@ -0,0 +1,21 @@
+{
+    "embeddings": {
+        "embedding_file": "model_components/glove.6B.200d.txt",
+        "dropout": 0.2
+    },
+    "classifier": {
+        "classes": [ "False", "True" ],
+        "has_query": 1,
+        "hidden_size": 32,
+        "mlp_size": 128,
+        "dropout": 0.2,
+        "batch_size": 16,
+        "epochs": 50,
+        "attention_epochs": 50,
+        "patience": 10,
+        "lr": 1e-3,
+        "dropout": 0.2,
+        "k_fraction": 0.07,
+        "threshold": 0.1
+    }
+}