Upload 7 files

README.md

@@ -0,0 +1,118 @@
+# RoBERTa-Base Model for Temporal Information Extraction
+
+This repository hosts a fine-tuned version of RoBERTa for **temporal information extraction**, where the model identifies and extracts time-related expressions (e.g., dates, durations) from text. The pipeline includes preprocessing, fine-tuning, and inference on labeled temporal datasets.
+
+---
+
+## Model Details
+
+- **Model Name:** RoBERTa-Base  
+- **Model Architecture:** RoBERTa Token Classification  
+- **Task:** Temporal Entity Extraction  
+- **Dataset:** Custom JSON format with annotated temporal SPO triples  
+- **Fine-tuning Framework:** Hugging Face Transformers  
+- **Output Labels:** `B-TIMEX`, `I-TIMEX`, `O`  
+
+---
+
+## Usage
+
+### Installation
+
+```bash
+pip install transformers datasets evaluate
+
+# Loading the Fine-Tuned Model
+
+from transformers import RobertaTokenizerFast, RobertaForTokenClassification
+import torch
+
+# Load model and tokenizer
+model = RobertaForTokenClassification.from_pretrained("./temporal_model")
+tokenizer = RobertaTokenizerFast.from_pretrained("./temporal_model", add_prefix_space=True)
+
+# Inference function
+def extract_temporal_entities(text):
+    tokens = text.split()
+    inputs = tokenizer(tokens, return_tensors="pt", is_split_into_words=True)
+    outputs = model(**inputs)
+    predictions = outputs.logits.argmax(dim=-1).squeeze().tolist()
+    word_ids = inputs.word_ids()[0]
+
+    temporal_spans = []
+    current = []
+    for idx, word_idx in enumerate(word_ids):
+        if word_idx is None:
+            continue
+        label = id2label[predictions[idx]]
+        if label == "B-TIMEX":
+            if current:
+                temporal_spans.append(" ".join(current))
+            current = [tokens[word_idx]]
+        elif label == "I-TIMEX":
+            current.append(tokens[word_idx])
+        else:
+            if current:
+                temporal_spans.append(" ".join(current))
+                current = []
+    if current:
+        temporal_spans.append(" ".join(current))
+    return temporal_spans
+
+
+# Performance Metrics
+Evaluation Accuracy: ~0.76
+
+F1 Score: Tracked using seqeval (BIO format)
+
+Evaluation Strategy: epoch
+
+# Fine-Tuning Details
+Dataset
+The dataset consists of manually or script-labeled SPO-style JSON entries with the following fields:
+
+text: Raw input string
+
+spo_list: A list of subject-predicate-object relations, including:
+
+Subject & Object Span
+
+Type (e.g., Date, Location)
+
+The text is tokenized, and BIO labels are applied for token classification.
+
+# Training Configuration
+
+Epochs: 3
+
+Batch Size: 16
+
+Learning Rate: 2e-5
+
+Max Sequence Length: 128 tokens
+
+Tokenizer: roberta-base with add_prefix_space=True
+
+# Repository Structure
+pgsql
+Copy
+Edit
+.
+├── temporal_model/             # Fine-tuned model and tokenizer
+│   ├── config.json
+│   ├── pytorch_model.bin
+│   ├── tokenizer_config.json
+│   ├── vocab.json
+│   └── special_tokens_map.json
+├── temporal-information-extraction.ipynb
+├── README.md
+
+# Limitations
+
+The model is domain-specific; generalization to other types of temporal expressions (e.g., informal text) may require additional training.
+
+BIO tagging may fail in overlapping or nested entity scenarios.
+
+# Contributing
+Contributions are welcome! Feel free to open an issue or submit a pull request to improve model performance or add new datasets.
+

config (1).json

@@ -0,0 +1,36 @@
+{
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float16",
+  "transformers_version": "4.51.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

model (1).safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0007fb807b07d163f3b7b1782b6c3aae9829d594abd477327ddfab522052db82
+size 248138558

special_tokens_map (1).json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config (1).json

@@ -0,0 +1,65 @@
+{
+  "add_prefix_space": true,
+  "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>",
+  "max_length": 128,
+  "model_max_length": 512,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "stride": 0,
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}