Save the first version of the model

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,15 @@
+{
+  "architectures": [
+    "AutextificationMTLModel"
+  ],
+  "auto_map": {
+    "AutoConfig": "config.AutextificationMTLConfig",
+    "AutoModel": "model.AutextificationMTLModel"
+  },
+  "hidden_nodes": 64,
+  "model_type": "custom-text-classifier",
+  "threshold": 0.9919,
+  "torch_dtype": "float32",
+  "transformer_name": "xlm-roberta-base",
+  "transformers_version": "4.25.1"
+}

config.py

@@ -0,0 +1,22 @@
+from transformers import PretrainedConfig
+
+
+class AutextificationMTLConfig(PretrainedConfig):
+    model_type = "custom-text-classifier"
+
+    def __init__(
+        self,
+        transformer_name: str = "xlm-roberta-base",
+        hidden_nodes: int = 64,
+        threshold: float = 0.9919,
+        **kwargs,
+    ):
+        if hidden_nodes <= 0:
+            raise ValueError(
+                f"`hidden_size` must be a positive number, got {hidden_nodes}."
+            )
+
+        self.transformer_name = transformer_name
+        self.hidden_nodes = hidden_nodes
+        self.threshold = threshold
+        super().__init__(**kwargs)

model.py

@@ -0,0 +1,39 @@
+import torch
+import torch.nn.functional as F
+from transformers import AutoModel, PreTrainedModel
+
+from .config import AutextificationMTLConfig
+
+
+class AutextificationMTLModel(PreTrainedModel):
+    config_class = AutextificationMTLConfig
+
+    def __init__(self, config: AutextificationMTLConfig):
+        super().__init__(config)
+
+        self.encoder = AutoModel.from_pretrained(config.transformer_name)
+        embedding_size = self.encoder.config.hidden_size
+        self.hidden = torch.nn.Linear(embedding_size, config.hidden_nodes)
+        self.out_generated = torch.nn.Linear(config.hidden_nodes, 1)
+        self.out_language = torch.nn.Linear(config.hidden_nodes, 1)
+
+        self.threshold = config.threshold
+
+    def forward(self, tensor):
+        output = self.encoder(
+            input_ids=tensor["input_ids"],
+            attention_mask=tensor["attention_mask"],
+            return_dict=True
+        )
+        pooler_output = output["pooler_output"]
+
+        out = F.relu(self.hidden(pooler_output))
+        out_generated = torch.sigmoid(self.out_generated(out))
+        out_language = torch.sigmoid(self.out_language(out))
+        out_verdict = out_generated > self.threshold
+
+        return {
+            "is_bot": out_verdict,
+            "bot_prob": out_generated,
+            "english_prob": out_language,
+        }

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:334a26950a52e078f3ae977657a385f1ad3355987fc1db67630882d9fa1e2155
+size 1112443781

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62c24cdc13d4c9952d63718d6c9fa4c287974249e16b7ade6d5a85e7bbb75626
+size 17082660

tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 512,
+  "name_or_path": "xlm-roberta-base",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "special_tokens_map_file": null,
+  "strip_accents": false,
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}