Upload 3 files

README.md

@@ -0,0 +1,98 @@
+---
+library_name: onnx
+pipeline_tag: feature-extraction
+license: mit
+tags:
+- eidora
+- eidora-model-zoo
+- onnx
+- onnxruntime
+- embeddings
+- text
+- e5
+- compute:light
+- modality:text
+base_model: intfloat/e5-small-v2
+datasets:
+- intfloat/e5
+metrics:
+- cosine-similarity
+model-index:
+- name: E5_SMALL_384
+  results: []
+---
+
+# E5_SMALL_384
+
+E5_SMALL_384 is a light text embedding model for fast semantic grouping and search in EIDORA. Choose it when you want a compact text model that runs comfortably on ordinary laptops.
+
+## Best For
+
+- Fast first-pass grouping of text notes, captions, and metadata.
+- Semantic search over medium and large text projects on laptops.
+- A compact starter model for EIDORA text embedding workflows.
+
+## Not Ideal For
+
+- Long-document reasoning or generation.
+- Fine-grained domain retrieval where a larger text embedding model is acceptable.
+- Image, video, or audio inputs.
+
+## Compute Tier
+
+Light: small download, low memory, faster CPU runtime. Intended for laptop CPU use and large first-pass projects.
+
+## Inputs
+
+- `text`: required text input from `text_source`.
+
+## Output
+
+The primary output is `embedding`, a float32 tensor shaped `[batch, 384]`. Embeddings are already normalized and are intended for cosine similarity.
+
+## Usage In EIDORA
+
+EIDORA shows this package as a light text embedding model in the Model Zoo. Use it for discovery maps, grouping, retrieval, and related embedding workflows.
+
+## Preprocessing
+
+- `text`: .
+
+## Authorship And Citation
+
+This ONNX package was produced by EIDORA from the original E5 Small v2 model. EIDORA converted the model to ONNX and is not the original model creator. Please cite Text Embeddings by Weakly-Supervised Contrastive Pre-training and the original model repository when using this converted model.
+
+Original model: https://github.com/microsoft/unilm/tree/master/e5
+
+Original paper: https://arxiv.org/abs/2212.03533
+
+Authors: Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, Furu Wei
+
+```bibtex
+@article{wang2022text,
+  title={Text Embeddings by Weakly-Supervised Contrastive Pre-training},
+  author={Wang, Liang and Yang, Nan and Huang, Xiaolong and Jiao, Binxing and Yang, Linjun and Jiang, Daxin and Majumder, Rangan and Wei, Furu},
+  journal={arXiv preprint arXiv:2212.03533},
+  year={2022}
+}
+```
+
+## Training Data And Provenance
+
+Base model: `intfloat/e5-small-v2`. Source repository: https://huggingface.co/intfloat/e5-small-v2. Known training data: Weakly supervised text pairs from the E5 training recipe, including CCPairs and supervised fine-tuning data described by the upstream authors. Package payload size: 133782386 bytes.
+
+## Evaluation And Validation
+
+The package validation checks that the ONNX graph loads with ONNX Runtime CPU execution, runs the declared fixtures, returns finite float32 embeddings with the declared shape, and matches the artifact hash recorded in `config.yaml`.
+
+## Limitations And Safety
+
+E5 embeddings can reflect the biases and language coverage of the upstream training data. The upstream recipe recommends query and passage prefixes; quality may drop if text is passed without the expected prefix style.
+
+## License And Attribution
+
+This package uses license `mit`. Upstream license: MIT. Converted to ONNX for EIDORA from the upstream intfloat E5 small v2 model.
+
+## Version
+
+Package version: 1.0.0. ONNX opset: 17. Exporter: eidora-onnx-exporter 0.1.0.

config.yaml

@@ -0,0 +1,150 @@
+schema_version: 1
+id: EIDORA/E5_small
+name: E5_small
+version: 1.0.0
+model_family: E5
+backend: onnx
+file_size: 134 MB
+runtime:
+  adapter: onnx_text
+  execution_provider: CPUExecutionProvider
+  model_path: model.onnx
+  input_names:
+  - input_ids
+  - attention_mask
+  output_name: embedding
+  onnxruntime:
+    opset: 17
+    tested_versions: '>=1.17,<2'
+artifact:
+  path: model.onnx
+  sha256: 67c0e5500336f67e22d59f942604308aca6af009ebc91ece04c649f86e1d2961
+  package_size_bytes: 133782386
+inputs:
+- id: text
+  modality: text
+  label: Text
+  required: true
+  source_kind: text_source
+  requirements:
+    max_tokens: 512
+    recommended_prefixes:
+    - 'query: '
+    - 'passage: '
+preprocess:
+  text:
+    tokenizer:
+      path: tokenizer
+      max_length: 512
+      truncation: true
+      padding: max_length
+    prefix_policy:
+      query: 'query: '
+      document: 'passage: '
+embedding:
+  dimensions: 384
+  feature_type: embedding
+  pooling: mean
+  normalized: true
+  similarity: cosine
+  output_name: embedding
+  dtype: float32
+  shape:
+  - batch
+  - 384
+display:
+  summary: 'Light: compact text embeddings for fast search, grouping, and discovery
+    on laptops.'
+  compute_tier: light
+  modality_labels:
+  - text
+  recommended_batch_size: 16
+validation:
+  fixtures:
+  - id: text_tokens_001
+    input_shapes:
+      input_ids:
+      - 1
+      - 16
+      attention_mask:
+      - 1
+      - 16
+    input_dtypes:
+      input_ids: int64
+      attention_mask: int64
+    expected_shape:
+    - 1
+    - 384
+    seed: 23
+  checks:
+    load_with: onnxruntime
+    execution_provider: CPUExecutionProvider
+    output_dtype: float32
+    finite: true
+    normalized_l2_range:
+    - 0.99
+    - 1.01
+provenance:
+  base_model: intfloat/e5-small-v2
+  source_repository: https://huggingface.co/intfloat/e5-small-v2
+  original_model_name: E5 Small v2
+  original_model_url: https://github.com/microsoft/unilm/tree/master/e5
+  authors:
+  - Liang Wang
+  - Nan Yang
+  - Xiaolong Huang
+  - Binxing Jiao
+  - Linjun Yang
+  - Daxin Jiang
+  - Rangan Majumder
+  - Furu Wei
+  paper_title: Text Embeddings by Weakly-Supervised Contrastive Pre-training
+  paper_url: https://arxiv.org/abs/2212.03533
+  upstream_license: MIT
+  training_data: Weakly supervised text pairs from the E5 training recipe, including
+    CCPairs and supervised fine-tuning data described by the upstream authors.
+  citation: "@article{wang2022text,\n  title={Text Embeddings by Weakly-Supervised\
+    \ Contrastive Pre-training},\n  author={Wang, Liang and Yang, Nan and Huang, Xiaolong\
+    \ and Jiao, Binxing and Yang, Linjun and Jiang, Daxin and Majumder, Rangan and\
+    \ Wei, Furu},\n  journal={arXiv preprint arXiv:2212.03533},\n  year={2022}\n}\n"
+  conversion_note: EIDORA produced this ONNX conversion and is not the original model
+    creator.
+  export_date: '2026-07-14'
+  exporter_version: eidora-onnx-exporter 0.1.0
+model_card:
+  best_for:
+  - Fast first-pass grouping of text notes, captions, and metadata.
+  - Semantic search over medium and large text projects on laptops.
+  - A compact starter model for EIDORA text embedding workflows.
+  not_ideal_for:
+  - Long-document reasoning or generation.
+  - Fine-grained domain retrieval where a larger text embedding model is acceptable.
+  - Image, video, or audio inputs.
+  limitations: E5 embeddings can reflect the biases and language coverage of the upstream
+    training data. The upstream recipe recommends query and passage prefixes; quality
+    may drop if text is passed without the expected prefix style.
+  license:
+    id: mit
+    attribution: Converted to ONNX for EIDORA from the upstream intfloat E5 small
+      v2 model.
+  huggingface:
+    org: eidora
+    repo_name: E5_SMALL_384
+    pipeline_tag: feature-extraction
+    tags:
+    - eidora
+    - eidora-model-zoo
+    - onnx
+    - onnxruntime
+    - embeddings
+    - text
+    - e5
+    - compute:light
+    - modality:text
+    datasets:
+    - intfloat/e5
+    metrics:
+    - cosine-similarity
+tokenizer:
+  path: tokenizer
+  source: intfloat/e5-small-v2

model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67c0e5500336f67e22d59f942604308aca6af009ebc91ece04c649f86e1d2961
+size 133070104