Add SetFit model

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": true,
+  "pooling_mode_mean_tokens": false,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false,
+  "include_prompt": true
+}

README.md

@@ -0,0 +1,302 @@
+---
+tags:
+- setfit
+- sentence-transformers
+- text-classification
+- generated_from_setfit_trainer
+widget:
+- text: ethiopia flood jul 2010 flood event lasted unknown ercs branch located north
+    east country reported 4 000 family affected flood 2 221 displaced temporarily
+    sheltered public school building 3 206 family reported affected flood 1 565 displaced
+    amhara region report indicated 800 family affected displaced flood afar region
+    total number affected family reported field 9 000 however number affected people
+    increasing due continuous torrential rain part country recently 5 000 family reported
+    displaced amhara tigrey afar region far due flooding occurred 22 24 august 2010
+    ercs icrc joint assessment tigrey amhara report ambasel tewlerda woredas south
+    wollo approx 1 368 hectare land crop flooded damaged hail storm based assessment
+    report approximately 3 745 hectare agricultural land flooded last week several
+    landslide reported field including 22 august 2010 mersa worgessa word north wollo
+    causing injury 19 death 5 people ifrc sep 2010 ethiopia
+- text: malaysia flood nov 2024 flood event lasted unknown end november 2024 malaysia
+    experienced heavy rainfall attributed northeast monsoon resulting escalating flooding
+    across nine state kelantan terengganu kedah pahang negeri sembilan johor perak
+    melaka perlis heavy rain caused significant damage livelihood house livestock
+    severely impacting affected community 2 december 2024 national disaster management
+    agency nadma reported approximately 137 410 people affected ongoing flood across
+    multiple area malaysia deputy prime minister informed medium year flooding worst
+    since 2014 kelantan terengganu particularly badly affected since 27 november total
+    633 temporary shelter center opened accommodate 40 922 family displaced flood
+    disaster claimed five life kelantan terengganu confirmed department social welfare
+    jkm ministry agriculture food security reported malaysia suffered approximately
+    chf 1 79 million loss due destruction rice paddy plantation caused flood significant
+    damage forced country increase reliance imported rice meet domestic need overall
+    malaysian agriculture sector face total estimated loss chf 3 77 million due disaster
+    malaysian meteorological department met malaysia forecasted continued adverse
+    weather condition including thunderstorm heavy rain strong wind across peninsular
+    malaysia 6 9 december 2024 condition expected exacerbate ongoing flooding increasing
+    number affected individual intensifying challenge emergency response recovery
+    effort persistent heavy rainfall already caused river water level surpass designated
+    danger threshold posing severe risk river overflow could inundate surrounding
+    area relentless rainfall caused extensive damage home also critical infrastructure
+    road airport railway particularly east coast state severely affected cutting intercity
+    connection complicating relief effort combined impact flood landslide underscore
+    urgent need enhanced mitigation measure coordinated response strategy ifrc 08
+    dec 2024 peninsular malaysia including johor kelantan pahang perak terengganu
+    state continues experience heavy rainfall consequent flood resulted displacement
+    damage according asean disaster information network adinet past day 6 517 people
+    displaced 44 evacuation centre across aforementioned state echo 12 dec 2024 4
+    january 2025 malaysia still grappling severe flooding caused ongoing northeast
+    monsoon began november 2024 expected persist march 2025 eastern coastal state
+    kelantan terengganu pahang johor hardest hit heavy rainfall leading widespread
+    flooding displacement significant disruption daily life metmalaysia forecast additional
+    five seven episode heavy rainfall monsoon season signalling situation may continue
+    several month flood caused substantial damage home infrastructure livelihood road
+    airport railway particularly affected east coast state disrupted intercity connectivity
+    hampered relief effort landslide compounded crisis underscoring need stronger
+    disaster mitigation response strategy additionally ministry agriculture food security
+    reported approximately chf 1 79 million loss due destruction rice paddy plantation
+    exacerbating economic impact affected community flood affected nine state across
+    malaysia including kelantan terengganu kedah pahang negeri sembilan johor perak
+    melaka perlis satellite imagery unosat show terengganu kelantan kedah severely
+    impacted floodwaters initially covering approximately 11 000 km terengganu kelantan
+    affecting 120 000 people kedah flood impacted 1 3 million people across 268 km
+    significant damage cropland persists even water begin recede ifrc 9 jan 2025 heavy
+    rainfall affecting peninsular malaysia since 10 january causing flood resulted
+    population displacement damage according asean disaster information network adinet
+    report 12 january 3 844 people displaced 38 evacuation centre 3 779 people johor
+    34 perak 31 terengganu state southern peninsular malaysia echo 13 jan 2025 past
+    day sabah sarawak state located malaysian borneo experiencing heavy rainfall flood
+    resulted casualty damage according medium least five people died 7 500 people
+    evacuated 5 385 sarawak 2 240 people affected sabah state echo 30 jan 2025 severe
+    monsoon flood continue devastate sabah sarawak displacing thousand causing widespread
+    disruption since 28 january 2025 continuous heavy rainfall compounded high tide
+    northeast monsoon led rising water level road inundation landslide sarawak situation
+    worsened due collision extreme monsoon rain high tide triggering large scale evacuation
+    activation multiple relief center 31 january 2025 12 486 evacuee 3 648 family
+    relocated 62 temporary relief center pps sarawak bintulu remains severely impacted
+    district sheltering 5 885 evacuee 1 649 family followed serian 2 307 evacuee 709
+    family samarahan 2 005 evacuee 670 family significantly affected district include
+    sibu 1 163 evacuee 293 family miri 650 evacuee 172 family kuching 475 evacuee
+    153 family single evacuee recorded mukah miri continuous heavy rainfall triggered
+    major landslide resulting tragic loss five life ifrc 1 feb 2025 according nadma
+    flooding landslide sabah sarawak resulted 5 fatality miri district report 3 february
+    1500 hr utc 7 2 9k family 9 7k person remain displaced across 50 evacuation center
+    sarawak bintulu serian miri sibu samarahan mukah sabah tongod kinabatangan aha
+    centre 3 feb 2025 heavy rainfall continued affect eastern malaysia malaysian part
+    borneo island since 29 january causing flood landslide resulted casualty damage
+    according international federation red cross ifrc 4 february death toll stand
+    five fatality ifrc also report nearly 12 500 evacuated people 62 temporary relief
+    center across sarawak state addition around 5 200 evacuated people 33 temporary
+    relief center reported across sabah state ifrc 4 feb 2025 malaysia
+- text: paraguay flood apr 2015 flood event lasted unknown 4 apr 2015 severe storm
+    hit several town department concepcin northern paraguay affecting house crop farm
+    animal authority estimate 5 000 people affected begun response providing roofing
+    material food medical attention ocha processing application emergency fund support
+    authority response ocha 13 apr 2015 per request paraguayan government usaid channeled
+    50 000 adra support response govt 15 apr 2015 may 2015 heavy rain caused overflowing
+    several river affected community asuncion central department according weather
+    expert amount rain atypical although intensity volume short time 3 000 family
+    affected district ypan villeta ypacara luque mariano roque alonso villa hayes
+    capiat limpio yaguarn ocha 11 may 2015 june 2015 national emergency agency sen
+    reported around 9 602 family 48 000 people affected flooding paraguay river asuncion
+    6 000 family received assistance sen coordinate action asuncion municipal council
+    emergency disaster paho 16 jun 2015 early july number affected family 32 000 23
+    000 received assistance department hit flood alto paraguay boquern presidente
+    hayes concepcin san pedro cordillera central guair caazap misiones eembuc government
+    paraguay 6 jul 2015 end july 2015 nearly 35 000 people affected flooding heavy
+    rain week stay shelter total 6 987 family asuncion shelter paho 24 jul 2015 last
+    week august heavy rain strong wind hail left 900 house affected department paraguar
+    san pedro cordillera central gov paraguay 28 aug 2015 paraguay
+- text: viet nam storm rai storm surge viet nam storm rai event lasted unknown afternoon
+    december 16 storm rai got stronger became super typhoon 19h 16 12 center super
+    typhoon central philippine wind level 16 gust level 17 moving northwest direction
+    speed 25 30km h past 6 hour intensity storm decreased one level longer level super
+    typhoon 01 17 12 center storm right central philippine wind level 15 gust level
+    17 philippine mobilized 54 response team evacuate 198 000 people prepared 26 million
+    414 000 food package respond storm currently human damage recorded viet nam
+- text: occupied palestinian territory cold wave dec 2013 cold wave event lasted unknown
+    announced heavy rain fall snow storm hit west bank gaza 10 december 2013 still
+    affecting palestinian population west bank palestine heavy rain snow generated
+    flood several part palestine thousand family evacuated house extreme weather condition
+    also caused several death including baby gaza reported dead family home inundated
+    ifrc 16 dec 2013 useful link ocha opt winter storm online system palestinian red
+    crescent society occupied palestinian territory
+metrics:
+- accuracy
+pipeline_tag: text-classification
+library_name: setfit
+inference: false
+base_model: avsolatorio/GIST-Embedding-v0
+model-index:
+- name: SetFit with avsolatorio/GIST-Embedding-v0
+  results:
+  - task:
+      type: text-classification
+      name: Text Classification
+    dataset:
+      name: Unknown
+      type: unknown
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.5272727272727272
+      name: Accuracy
+---
+
+# SetFit with avsolatorio/GIST-Embedding-v0
+
+This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [avsolatorio/GIST-Embedding-v0](https://huggingface.co/avsolatorio/GIST-Embedding-v0) as the Sentence Transformer embedding model. A [SetFitHead](huggingface.co/docs/setfit/reference/main#setfit.SetFitHead) instance is used for classification.
+
+The model has been trained using an efficient few-shot learning technique that involves:
+
+1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
+2. Training a classification head with features from the fine-tuned Sentence Transformer.
+
+## Model Details
+
+### Model Description
+- **Model Type:** SetFit
+- **Sentence Transformer body:** [avsolatorio/GIST-Embedding-v0](https://huggingface.co/avsolatorio/GIST-Embedding-v0)
+- **Classification head:** a [SetFitHead](huggingface.co/docs/setfit/reference/main#setfit.SetFitHead) instance
+- **Maximum Sequence Length:** 512 tokens
+<!-- - **Number of Classes:** Unknown -->
+<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
+- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
+- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
+
+## Evaluation
+
+### Metrics
+| Label   | Accuracy |
+|:--------|:---------|
+| **all** | 0.5273   |
+
+## Uses
+
+### Direct Use for Inference
+
+First install the SetFit library:
+
+```bash
+pip install setfit
+```
+
+Then you can load this model and run inference.
+
+```python
+from setfit import SetFitModel
+
+# Download from the 🤗 Hub
+model = SetFitModel.from_pretrained("AlexBayer/GIST_SetFit_HIPs_v1")
+# Run inference
+preds = model("occupied palestinian territory cold wave dec 2013 cold wave event lasted unknown announced heavy rain fall snow storm hit west bank gaza 10 december 2013 still affecting palestinian population west bank palestine heavy rain snow generated flood several part palestine thousand family evacuated house extreme weather condition also caused several death including baby gaza reported dead family home inundated ifrc 16 dec 2013 useful link ocha opt winter storm online system palestinian red crescent society occupied palestinian territory")
+```
+
+<!--
+### Downstream Use
+
+*List how someone could finetune this model on their own dataset.*
+-->
+
+<!--
+### Out-of-Scope Use
+
+*List how the model may foreseeably be misused and address what users ought not to do with the model.*
+-->
+
+<!--
+## Bias, Risks and Limitations
+
+*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
+-->
+
+<!--
+### Recommendations
+
+*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
+-->
+
+## Training Details
+
+### Training Set Metrics
+| Training set | Min | Median   | Max  |
+|:-------------|:----|:---------|:-----|
+| Word count   | 34  | 319.4125 | 2470 |
+
+### Training Hyperparameters
+- batch_size: (16, 2)
+- num_epochs: (1, 16)
+- max_steps: -1
+- sampling_strategy: undersampling
+- body_learning_rate: (2.658312040445757e-05, 2.395494892138971e-05)
+- head_learning_rate: 0.01770304761261279
+- loss: CosineSimilarityLoss
+- distance_metric: cosine_distance
+- margin: 0.25
+- end_to_end: True
+- use_amp: True
+- warmup_proportion: 0.1
+- l2_weight: 0.05
+- max_length: 512
+- seed: 42
+- eval_max_steps: -1
+- load_best_model_at_end: False
+
+### Training Results
+| Epoch  | Step | Training Loss | Validation Loss |
+|:------:|:----:|:-------------:|:---------------:|
+| 0.1534 | 25   | 0.2424        | -               |
+| 0.3067 | 50   | 0.1673        | -               |
+| 0.4601 | 75   | 0.1422        | -               |
+| 0.6135 | 100  | 0.1242        | -               |
+| 0.7669 | 125  | 0.1148        | -               |
+| 0.9202 | 150  | 0.096         | -               |
+
+### Framework Versions
+- Python: 3.11.12
+- SetFit: 1.1.2
+- Sentence Transformers: 3.4.1
+- Transformers: 4.51.3
+- PyTorch: 2.6.0+cu124
+- Datasets: 3.5.1
+- Tokenizers: 0.21.1
+
+## Citation
+
+### BibTeX
+```bibtex
+@article{https://doi.org/10.48550/arxiv.2209.11055,
+    doi = {10.48550/ARXIV.2209.11055},
+    url = {https://arxiv.org/abs/2209.11055},
+    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
+    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
+    title = {Efficient Few-Shot Learning Without Prompts},
+    publisher = {arXiv},
+    year = {2022},
+    copyright = {Creative Commons Attribution 4.0 International}
+}
+```
+
+<!--
+## Glossary
+
+*Clearly define terms in order to be accessible across audiences.*
+-->
+
+<!--
+## Model Card Authors
+
+*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
+-->
+
+<!--
+## Model Card Contact
+
+*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
+-->

config.json

@@ -0,0 +1,31 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

config_sentence_transformers.json

@@ -0,0 +1,10 @@
+{
+  "__version__": {
+    "sentence_transformers": "3.4.1",
+    "transformers": "4.51.3",
+    "pytorch": "2.6.0+cu124"
+  },
+  "prompts": {},
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

config_setfit.json

@@ -0,0 +1,4 @@
+{
+  "labels": null,
+  "normalize_embeddings": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:642c5e314d0044d60685836bc30b40cfe77c43b4d0c979dd6a00861c2d02cbdc
+size 437951328

model_head.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fb62f243a9f75c09e6777f9cdaf63a577d0a6d9ead34fac6e46fbb7f9fe830f
+size 44235

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": true
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}