Upload folder using huggingface_hub

config.json

@@ -0,0 +1,67 @@
+{
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "CSR/Brand",
+    "1": "Deal",
+    "2": "Dividend",
+    "3": "Employment",
+    "4": "Expense",
+    "5": "Facility",
+    "6": "FinancialReport",
+    "7": "Financing",
+    "8": "Investment",
+    "9": "Legal",
+    "10": "Macroeconomics",
+    "11": "Merger/Acquisition",
+    "12": "Product/Service",
+    "13": "Profit/Loss",
+    "14": "Rating",
+    "15": "Revenue",
+    "16": "SalesVolume",
+    "17": "SecurityValue"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "CSR/Brand": 0,
+    "Deal": 1,
+    "Dividend": 2,
+    "Employment": 3,
+    "Expense": 4,
+    "Facility": 5,
+    "FinancialReport": 6,
+    "Financing": 7,
+    "Investment": 8,
+    "Legal": 9,
+    "Macroeconomics": 10,
+    "Merger/Acquisition": 11,
+    "Product/Service": 12,
+    "Profit/Loss": 13,
+    "Rating": 14,
+    "Revenue": 15,
+    "SalesVolume": 16,
+    "SecurityValue": 17
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "multi_label_classification",
+  "transformers_version": "4.57.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

handler.py

@@ -0,0 +1,54 @@
+import torch
+import numpy as np
+from typing import Dict, List, Any
+
+class EndpointHandler:
+    def __init__(self, path=""):
+        # Load the model and tokenizer
+        from transformers import AutoModelForSequenceClassification, AutoTokenizer
+        self.model = AutoModelForSequenceClassification.from_pretrained(path)
+        self.tokenizer = AutoTokenizer.from_pretrained(path)
+        
+        # Load per-class thresholds
+        thresholds_path = f"{path}/thresholds.npy"
+        self.thresholds = np.load(thresholds_path)
+        
+        self.model.eval()
+        
+    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
+        """
+        Args:
+            data (Dict[str, Any]): Input data containing 'inputs' key
+        Returns:
+            List[Dict[str, Any]]: Predictions with labels and scores
+        """
+        inputs_text = data.pop("inputs", data)
+        
+        # Tokenize
+        inputs = self.tokenizer(
+            inputs_text,
+            return_tensors="pt",
+            truncation=True,
+            padding="max_length",
+            max_length=128
+        )
+        
+        # Inference
+        with torch.no_grad():
+            outputs = self.model(**inputs)
+            logits = outputs.logits[0]
+            probs = torch.sigmoid(logits).cpu().numpy()
+        
+        # Apply per-class thresholds
+        predictions = []
+        for idx, prob in enumerate(probs):
+            if prob >= self.thresholds[idx]:
+                predictions.append({
+                    "label": self.model.config.id2label[idx],
+                    "score": float(prob)
+                })
+        
+        # Sort by score descending
+        predictions = sorted(predictions, key=lambda x: x["score"], reverse=True)
+        
+        return predictions

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2fea05419ed4f1c5d52f6d731e0aec523403703e35d23c689b45521de08ad8c
+size 1421561016

readme.md

@@ -0,0 +1,58 @@
+# RoBERTa Multi-Label Financial Event Classifier (SENTiVENT)
+
+This model is a fine-tuned **RoBERTa** classifier that predicts one or more **financial event types** from a news headline or short piece of text.
+
+Given a headline, the model outputs probabilities for multiple event categories such as mergers, earnings reports, legal actions, investments, and more. It is designed for **multi-label classification**, meaning a single headline can belong to multiple event types at once.
+
+---
+
+## Event Labels
+
+The model predicts the following 18 event categories:
+
+- CSR/Brand  
+- Deal  
+- Dividend  
+- Employment  
+- Expense  
+- Facility  
+- FinancialReport  
+- Financing  
+- Investment  
+- Legal  
+- Macroeconomics  
+- Merger/Acquisition  
+- Product/Service  
+- Profit/Loss  
+- Rating  
+- Revenue  
+- SalesVolume  
+- SecurityValue  
+
+---
+
+## Intended Use
+
+This model is useful for:
+
+- Financial news analysis
+- Market event extraction
+- Trading signal pipelines
+- Knowledge graph population
+- Research in finance-focused NLP
+
+It works best on **short financial news headlines or sentences**.
+
+---
+
+## Model Details
+
+- Base model: `roberta-base`
+- Task: Multi-label text classification
+- Activation: Sigmoid (per-label probabilities)
+- Loss: Binary Cross Entropy
+- Problem type: `multi_label_classification`
+
+The model configuration includes `id2label`, `label2id`, and the correct problem type so it works cleanly with Hugging Face pipelines.
+
+---

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>"
+}

thresholds.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae12d5e934f881025c5b757c8fe3b3222b03acb23d95742ba193178f7086c7bf
+size 418

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4777712ed9d7b32a340e8548f6cfa3388f82beba906b15f717b297b0eae2a82
+size 5777