mletsana commited on
Commit
82ab285
·
verified ·
1 Parent(s): 4b98021

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

Browse files
Files changed (5) hide show
  1. README.md +46 -0
  2. config.json +62 -0
  3. model.safetensors +3 -0
  4. tokenizer.json +0 -0
  5. tokenizer_config.json +21 -0
README.md ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ license: mit
4
+ library_name: pytorch
5
+ tags:
6
+ - climate-journalism
7
+ - cjis-copilot
8
+ - themes
9
+ - multi-label-classification
10
+ # Base: Custom MLM-trained DeBERTa (not on HuggingFace Hub)
11
+ ---
12
+
13
+ # CJIS Co-pilot: Themes Classifier
14
+
15
+ Part of the Climate Journalism Intelligence System (CJIS) v3.0 "Journalistic Co-pilot".
16
+
17
+ ## Architecture
18
+ - Base: Climate-pretrained DeBERTa (MLM on 20k climate articles)
19
+ - Head: Specialist classifier for themes
20
+
21
+ ## Task
22
+ Themes classification for climate journalism articles.
23
+
24
+ ## Labels (10)
25
+ - `emissions_carbon_pricing`
26
+ - `climate_finance_development`
27
+ - `food_security_public_health`
28
+ - `renewable_energy_tech`
29
+ - `fossil_fuel_industry`
30
+ - `grid_infrastructure`
31
+ - `just_transition`
32
+ - `climate_litigation`
33
+ - `extreme_weather_events`
34
+ - `biodiversity_ecosystems`
35
+
36
+ ## Usage
37
+ ```python
38
+ from transformers import pipeline
39
+
40
+ classifier = pipeline(
41
+ "text-classification",
42
+ model="maai-org/cjis-themes-classifier",
43
+ return_all_scores=True
44
+ )
45
+
46
+ result = classifier("Your climate article here...")
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaV2ForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": null,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "emissions_carbon_pricing",
14
+ "1": "climate_finance_development",
15
+ "2": "food_security_public_health",
16
+ "3": "renewable_energy_tech",
17
+ "4": "fossil_fuel_industry",
18
+ "5": "grid_infrastructure",
19
+ "6": "just_transition",
20
+ "7": "climate_litigation",
21
+ "8": "extreme_weather_events",
22
+ "9": "biodiversity_ecosystems"
23
+ },
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 3072,
26
+ "label2id": {
27
+ "biodiversity_ecosystems": 9,
28
+ "climate_finance_development": 1,
29
+ "climate_litigation": 7,
30
+ "emissions_carbon_pricing": 0,
31
+ "extreme_weather_events": 8,
32
+ "food_security_public_health": 2,
33
+ "fossil_fuel_industry": 4,
34
+ "grid_infrastructure": 5,
35
+ "just_transition": 6,
36
+ "renewable_energy_tech": 3
37
+ },
38
+ "layer_norm_eps": 1e-07,
39
+ "legacy": true,
40
+ "max_position_embeddings": 512,
41
+ "max_relative_positions": -1,
42
+ "model_type": "deberta-v2",
43
+ "norm_rel_ebd": "layer_norm",
44
+ "num_attention_heads": 12,
45
+ "num_hidden_layers": 12,
46
+ "pad_token_id": 0,
47
+ "pooler_dropout": 0,
48
+ "pooler_hidden_act": "gelu",
49
+ "pooler_hidden_size": 768,
50
+ "pos_att_type": [
51
+ "p2c",
52
+ "c2p"
53
+ ],
54
+ "position_biased_input": false,
55
+ "position_buckets": 256,
56
+ "relative_attention": true,
57
+ "share_att_key": true,
58
+ "tie_word_embeddings": true,
59
+ "transformers_version": "5.0.0",
60
+ "type_vocab_size": 0,
61
+ "vocab_size": 128100
62
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92fec02984a59a3228aeffade9bd53a8f965ecbe2ad7e7427bce3db1487b5dbb
3
+ size 737743888
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "clean_up_tokenization_spaces": false,
6
+ "cls_token": "[CLS]",
7
+ "do_lower_case": false,
8
+ "eos_token": "[SEP]",
9
+ "is_local": true,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 1000000000000000019884624838656,
12
+ "model_specific_special_tokens": {},
13
+ "pad_token": "[PAD]",
14
+ "sep_token": "[SEP]",
15
+ "sp_model_kwargs": {},
16
+ "split_by_punct": false,
17
+ "tokenizer_class": "DebertaV2Tokenizer",
18
+ "unk_id": 3,
19
+ "unk_token": "[UNK]",
20
+ "vocab_type": "spm"
21
+ }