add model

config.json

@@ -5,6 +5,9 @@
     "DistilBertForMultilabelSequenceClassification"
   ],
   "attention_dropout": 0.1,
+  "auto_map": {
+    "AutoModelForSequenceClassification": "distilbert_model.DistilBertForMultilabelSequenceClassification"
+  },
   "dim": 768,
   "dropout": 0.1,
   "hidden_dim": 3072,
@@ -25,19 +28,19 @@
   },
   "initializer_range": 0.02,
   "label2id": {
-    "admiration": "0",
-    "annoyance": "1",
-    "approval": "2",
-    "caring": "3",
-    "confusion": "4",
-    "curiosity": "5",
-    "disapproval": "6",
-    "excitement": "7",
-    "joy": "8",
-    "nervousness": "9",
-    "neutral": "12",
-    "optimism": "10",
-    "sadness": "11"
+    "admiration": 0,
+    "annoyance": 1,
+    "approval": 2,
+    "caring": 3,
+    "confusion": 4,
+    "curiosity": 5,
+    "disapproval": 6,
+    "excitement": 7,
+    "joy": 8,
+    "nervousness": 9,
+    "neutral": 12,
+    "optimism": 10,
+    "sadness": 11
   },
   "max_position_embeddings": 512,
   "model_type": "distilbert",
@@ -49,6 +52,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.12.5",
+  "transformers_version": "4.17.0",
   "vocab_size": 30522
 }

distilbert_model.py

@@ -0,0 +1,49 @@
+import torch
+import pandas as pd
+import numpy as np
+from pathlib import Path
+from transformers import DistilBertForSequenceClassification
+from transformers.modeling_outputs import SequenceClassifierOutput
+
+class DistilBertForMultilabelSequenceClassification(DistilBertForSequenceClassification):
+    def __init__(self, config):
+        super().__init__(config)
+
+    def forward(self,
+            input_ids=None,
+            attention_mask=None,
+            head_mask=None,
+            inputs_embeds=None,
+            labels=None,
+            output_attentions=None,
+            output_hidden_states=None,
+            return_dict=None):
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.distilbert(input_ids,
+                attention_mask=attention_mask,
+                head_mask=head_mask,
+                inputs_embeds=inputs_embeds,
+                output_attentions=output_attentions,
+                output_hidden_states=output_hidden_states,
+                return_dict=return_dict)
+
+        hidden_state = outputs[0]
+        pooled_output = hidden_state[:, 0]
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            loss_fct = torch.nn.BCEWithLogitsLoss()
+            loss = loss_fct(logits.view(-1, self.num_labels),
+                    labels.float().view(-1, self.num_labels))
+
+        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)

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2e439515aee872fbf4e392d59acb1c05cf662068c46bb9b8d5fc2dc01cdff1e
-size 267893873
+oid sha256:c0007ac949fda38934a5c5428f8553ba86505512dbfc116399e205553f8460a9
+size 267894321