CJIS v3.0 cjis-frames-classifier - Climate journalism multi-label classifier

README.md

@@ -0,0 +1,44 @@
+---
+language: en
+license: mit
+library_name: pytorch
+tags:
+- climate-journalism
+- cjis-copilot
+- frames
+- multi-label-classification
+# Base: Custom MLM-trained DeBERTa (not on HuggingFace Hub)
+---
+
+# CJIS Co-pilot: Frames Classifier
+
+Part of the Climate Journalism Intelligence System (CJIS) v3.0 "Journalistic Co-pilot".
+
+## Architecture
+- Base: Climate-pretrained DeBERTa (MLM on 20k climate articles)
+- Head: Specialist classifier for frames
+
+## Task
+Frames classification for climate journalism articles.
+
+## Labels (8)
+- `scientific`
+- `crisis`
+- `economic_solutions`
+- `political`
+- `justice_responsibility`
+- `health_ecological`
+- `security`
+- `human_interest`
+
+## Usage
+```python
+from transformers import pipeline
+
+classifier = pipeline(
+    "text-classification",
+    model="maai-org/cjis-frames-classifier",
+    return_all_scores=True
+)
+
+result = classifier("Your climate article here...")

config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "DebertaV2ForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": null,
+  "dtype": "float32",
+  "eos_token_id": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "scientific",
+    "1": "crisis",
+    "2": "economic_solutions",
+    "3": "political",
+    "4": "justice_responsibility",
+    "5": "health_ecological",
+    "6": "security",
+    "7": "human_interest"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "crisis": 1,
+    "economic_solutions": 2,
+    "health_ecological": 5,
+    "human_interest": 7,
+    "justice_responsibility": 4,
+    "political": 3,
+    "scientific": 0,
+    "security": 6
+  },
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.0.0",
+  "type_vocab_size": 0,
+  "vocab_size": 128100
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8190c0ba919ed4ff6fd7cba655ef1556bff761494a16a1f3b981f2a1ed9c5be
+size 737737736

tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "is_local": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "model_specific_special_tokens": {},
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "sp_model_kwargs": {},
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "unk_id": 3,
+  "unk_token": "[UNK]",
+  "vocab_type": "spm"
+}