Upload 11 files

README.md

@@ -1,3 +1,217 @@
----
-license: apache-2.0
----
+---
+library_name: transformers
+license: apache-2.0
+language:
+- en
+tags:
+- reranker
+- cross-encoder
+- transformers.js
+pipeline_tag: text-classification
+---
+
+<br><br>
+
+<p align="center">
+<img src="https://huggingface.co/datasets/jinaai/documentation-images/resolve/main/logo.webp" alt="Jina AI: Your Search Foundation, Supercharged!" width="150px">
+</p>
+
+<p align="center">
+<b>Trained by <a href="https://jina.ai/"><b>Jina AI</b></a>.</b>
+</p>
+
+# jina-reranker-v1-tiny-en
+
+This model is designed for **blazing-fast** reranking while maintaining **competitive performance**. What's more, it leverages the power of our [JinaBERT](https://arxiv.org/abs/2310.19923) model as its foundation. `JinaBERT` itself is a unique variant of the BERT architecture that supports the symmetric bidirectional variant of [ALiBi](https://arxiv.org/abs/2108.12409). This allows `jina-reranker-v1-tiny-en` to process significantly longer sequences of text compared to other reranking models, up to an impressive **8,192** tokens.
+
+To achieve the remarkable speed, the `jina-reranker-v1-tiny-en` employ a technique called knowledge distillation. Here, a complex, but slower, model (like our original [jina-reranker-v1-base-en](https://jina.ai/reranker/)) acts as a teacher, condensing its knowledge into a smaller, faster student model. This student retains most of the teacher's knowledge, allowing it to deliver similar accuracy in a fraction of the time.
+
+Here's a breakdown of the reranker models we provide:
+
+| Model Name                                                                           | Layers | Hidden Size | Parameters (Millions) |
+| ------------------------------------------------------------------------------------ | ------ | ----------- | --------------------- |
+| [jina-reranker-v1-base-en](https://jina.ai/reranker/)                                | 12     | 768         | 137.0                 |
+| [jina-reranker-v1-turbo-en](https://huggingface.co/jinaai/jina-reranker-v1-turbo-en) | 6      | 384         | 37.8                  |
+| [jina-reranker-v1-tiny-en](https://huggingface.co/jinaai/jina-reranker-v1-tiny-en)   | 4      | 384         | 33.0                  |
+
+> Currently, the `jina-reranker-v1-base-en` model is not available on Hugging Face. You can access it via the [Jina AI Reranker API](https://jina.ai/reranker/).
+
+As you can see, the `jina-reranker-v1-turbo-en` offers a balanced approach with **6 layers** and **37.8 million** parameters. This translates to fast search and reranking while preserving a high degree of accuracy. The `jina-reranker-v1-tiny-en` prioritizes speed even further, achieving the fastest inference speeds with its **4-layer**, **33.0 million** parameter architecture. This makes it ideal for scenarios where absolute top accuracy is less crucial.
+
+# Usage
+
+1. The easiest way to starting using `jina-reranker-v1-tiny-en` is to use Jina AI's [Reranker API](https://jina.ai/reranker/).
+
+```bash
+curl https://api.jina.ai/v1/rerank \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer YOUR_API_KEY" \
+  -d '{
+  "model": "jina-reranker-v1-tiny-en",
+  "query": "Organic skincare products for sensitive skin",
+  "documents": [
+    "Eco-friendly kitchenware for modern homes",
+    "Biodegradable cleaning supplies for eco-conscious consumers",
+    "Organic cotton baby clothes for sensitive skin",
+    "Natural organic skincare range for sensitive skin",
+    "Tech gadgets for smart homes: 2024 edition",
+    "Sustainable gardening tools and compost solutions",
+    "Sensitive skin-friendly facial cleansers and toners",
+    "Organic food wraps and storage solutions",
+    "All-natural pet food for dogs with allergies",
+    "Yoga mats made from recycled materials"
+  ],
+  "top_n": 3
+}'
+```
+
+2. Alternatively, you can use the latest version of the `sentence-transformers>=0.27.0` library. You can install it via pip:
+
+```bash
+pip install -U sentence-transformers
+```
+
+Then, you can use the following code to interact with the model:
+
+```python
+from sentence_transformers import CrossEncoder
+
+# Load the model, here we use our tiny sized model
+model = CrossEncoder("jinaai/jina-reranker-v1-tiny-en", trust_remote_code=True)
+
+# Example query and documents
+query = "Organic skincare products for sensitive skin"
+documents = [
+    "Eco-friendly kitchenware for modern homes",
+    "Biodegradable cleaning supplies for eco-conscious consumers",
+    "Organic cotton baby clothes for sensitive skin",
+    "Natural organic skincare range for sensitive skin",
+    "Tech gadgets for smart homes: 2024 edition",
+    "Sustainable gardening tools and compost solutions",
+    "Sensitive skin-friendly facial cleansers and toners",
+    "Organic food wraps and storage solutions",
+    "All-natural pet food for dogs with allergies",
+    "Yoga mats made from recycled materials"
+]
+
+results = model.rank(query, documents, return_documents=True, top_k=3)
+```
+
+3. You can also use the `transformers` library to interact with the model programmatically.
+
+```python
+!pip install transformers
+from transformers import AutoModelForSequenceClassification
+
+model = AutoModelForSequenceClassification.from_pretrained(
+    'jinaai/jina-reranker-v1-tiny-en', num_labels=1, trust_remote_code=True
+)
+
+# Example query and documents
+query = "Organic skincare products for sensitive skin"
+documents = [
+    "Eco-friendly kitchenware for modern homes",
+    "Biodegradable cleaning supplies for eco-conscious consumers",
+    "Organic cotton baby clothes for sensitive skin",
+    "Natural organic skincare range for sensitive skin",
+    "Tech gadgets for smart homes: 2024 edition",
+    "Sustainable gardening tools and compost solutions",
+    "Sensitive skin-friendly facial cleansers and toners",
+    "Organic food wraps and storage solutions",
+    "All-natural pet food for dogs with allergies",
+    "Yoga mats made from recycled materials"
+]
+
+# construct sentence pairs
+sentence_pairs = [[query, doc] for doc in documents]
+
+scores = model.compute_score(sentence_pairs)
+```
+
+4. You can also use the `transformers.js` library to run the model directly in JavaScript (in-browser, Node.js, Deno, etc.)!
+
+If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@xenova/transformers) using:
+```bash
+npm i @xenova/transformers
+```
+
+Then, you can use the following code to interact with the model:
+```js
+import { AutoTokenizer, AutoModelForSequenceClassification } from '@xenova/transformers';
+
+const model_id = 'jinaai/jina-reranker-v1-tiny-en';
+const model = await AutoModelForSequenceClassification.from_pretrained(model_id, { quantized: false });
+const tokenizer = await AutoTokenizer.from_pretrained(model_id);
+
+/**
+ * Performs ranking with the CrossEncoder on the given query and documents. Returns a sorted list with the document indices and scores.
+ * @param {string} query A single query
+ * @param {string[]} documents A list of documents
+ * @param {Object} options Options for ranking
+ * @param {number} [options.top_k=undefined] Return the top-k documents. If undefined, all documents are returned.
+ * @param {number} [options.return_documents=false] If true, also returns the documents. If false, only returns the indices and scores.
+ */
+async function rank(query, documents, {
+    top_k = undefined,
+    return_documents = false,
+} = {}) {
+    const inputs = tokenizer(
+        new Array(documents.length).fill(query),
+        { text_pair: documents, padding: true, truncation: true }
+    )
+    const { logits } = await model(inputs);
+    return logits.sigmoid().tolist()
+        .map(([score], i) => ({
+            corpus_id: i,
+            score,
+            ...(return_documents ? { text: documents[i] } : {})
+        })).sort((a, b) => b.score - a.score).slice(0, top_k);
+}
+
+// Example usage:
+const query = "Organic skincare products for sensitive skin"
+const documents = [
+    "Eco-friendly kitchenware for modern homes",
+    "Biodegradable cleaning supplies for eco-conscious consumers",
+    "Organic cotton baby clothes for sensitive skin",
+    "Natural organic skincare range for sensitive skin",
+    "Tech gadgets for smart homes: 2024 edition",
+    "Sustainable gardening tools and compost solutions",
+    "Sensitive skin-friendly facial cleansers and toners",
+    "Organic food wraps and storage solutions",
+    "All-natural pet food for dogs with allergies",
+    "Yoga mats made from recycled materials",
+]
+
+const results = await rank(query, documents, { return_documents: true, top_k: 3 });
+console.log(results);
+```
+
+
+That's it! You can now use the `jina-reranker-v1-tiny-en` model in your projects.
+
+# Evaluation
+
+We evaluated Jina Reranker on 3 key benchmarks to ensure top-tier performance and search relevance.
+
+| Model Name                                 | NDCG@10 (17 BEIR datasets) | NDCG@10 (5 LoCo datasets) | Hit Rate (LlamaIndex RAG) |
+| ------------------------------------------ | -------------------------- | ------------------------- | ------------------------- |
+| `jina-reranker-v1-base-en`                 | **52.45**                  | **87.31**                 | **85.53**                 |
+| `jina-reranker-v1-turbo-en`                | **49.60**                  | **69.21**                 | **85.13**                 |
+| `jina-reranker-v1-tiny-en` (you are here)  | **48.54**                  | **70.29**                 | **85.00**                 |
+| `mxbai-rerank-base-v1`                     | 49.19                      | -                         | 82.50                     |
+| `mxbai-rerank-xsmall-v1`                   | 48.80                      | -                         | 83.69                     |
+| `ms-marco-MiniLM-L-6-v2`                   | 48.64                      | -                         | 82.63                     |
+| `ms-marco-MiniLM-L-4-v2`                   | 47.81                      | -                         | 83.82                     |
+| `bge-reranker-base`                        | 47.89                      | -                         | 83.03                     |
+
+**Note:**
+
+- `NDCG@10` is a measure of ranking quality, with higher scores indicating better search results. `Hit Rate` measures the percentage of relevant documents that appear in the top 10 search results.
+- The results of LoCo datasets on other models are not available since they **do not support** long documents more than 512 tokens.
+
+For more details, please refer to our [benchmarking sheets](https://docs.google.com/spreadsheets/d/1V8pZjENdBBqrKMzZzOWc2aL60wtnR0yrEBY3urfO5P4/edit?usp=sharing).
+
+# Contact
+
+Join our [Discord community](https://discord.jina.ai/) and chat with other community members about ideas.

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "jinaai/jina-bert-implementation",
+  "architectures": ["JinaBertModel"],
+  "attention_probs_dropout_prob": 0.1,
+  "attn_implementation": "torch",
+  "auto_map": {
+    "AutoConfig": "configuration_bert.JinaBertConfig",
+    "AutoModel": "modeling_bert.JinaBertModel",
+    "AutoModelForMaskedLM": "modeling_bert.JinaBertForMaskedLM",
+    "AutoModelForQuestionAnswering": "modeling_bert.JinaBertForQuestionAnswering",
+    "AutoModelForSequenceClassification": "modeling_bert.JinaBertForSequenceClassification",
+    "AutoModelForTokenClassification": "modeling_bert.JinaBertForTokenClassification"
+  },
+  "classifier_dropout": null,
+  "emb_pooler": "mean",
+  "feed_forward_type": "geglu",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 8192,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 4,
+  "pad_token_id": 0,
+  "position_embedding_type": "alibi",
+  "torch_dtype": "float16",
+  "transformers_version": "4.30.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 61056
+}

configuration_bert.py

@@ -0,0 +1,168 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+# Copyright (c) 2023 Jina AI GmbH. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" BERT model configuration"""
+from collections import OrderedDict
+from typing import Mapping
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.onnx import OnnxConfig
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+
+class JinaBertConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`JinaBertModel`]. It is used to
+    instantiate a BERT model according to the specified arguments, defining the model architecture. Instantiating a
+    configuration with the defaults will yield a similar configuration to that of the BERT
+    [bert-base-uncased](https://huggingface.co/bert-base-uncased) architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 30522):
+            Vocabulary size of the BERT model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`BertModel`] or [`TFBertModel`].
+        hidden_size (`int`, *optional*, defaults to 768):
+            Dimensionality of the encoder layers and the pooler layer.
+        num_hidden_layers (`int`, *optional*, defaults to 12):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (`int`, *optional*, defaults to 12):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        intermediate_size (`int`, *optional*, defaults to 3072):
+            Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
+        hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+            `"relu"`, `"silu"` and `"gelu_new"` are supported.
+        hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
+            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+        attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
+            The dropout ratio for the attention probabilities.
+        max_position_embeddings (`int`, *optional*, defaults to 512):
+            The maximum sequence length that this model might ever be used with. Typically set this to something large
+            just in case (e.g., 512 or 1024 or 2048).
+        type_vocab_size (`int`, *optional*, defaults to 2):
+            The vocabulary size of the `token_type_ids` passed when calling [`BertModel`] or [`TFBertModel`].
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+            The epsilon used by the layer normalization layers.
+        position_embedding_type (`str`, *optional*, defaults to `"absolute"`):
+            Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For
+            positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to
+            [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155).
+            For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models
+            with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658).
+        is_decoder (`bool`, *optional*, defaults to `False`):
+            Whether the model is used as a decoder or not. If `False`, the model is used as an encoder.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        classifier_dropout (`float`, *optional*):
+            The dropout ratio for the classification head.
+        feed_forward_type (`str`, *optional*, defaults to `"original"`):
+            The type of feed forward layer to use in the bert layers.
+            Can be one of GLU variants, e.g. `"reglu"`, `"geglu"`
+        emb_pooler (`str`, *optional*, defaults to `None`):
+            The function to use for pooling the last layer embeddings to get the sentence embeddings.
+            Should be one of `None`, `"mean"`.
+        attn_implementation (`str`, *optional*, defaults to `"torch"`):
+            The implementation of the self-attention layer. Can be one of:
+            - `None` for the original implementation,
+            - `torch` for the PyTorch SDPA implementation,
+
+    Examples:
+
+    ```python
+    >>> from transformers import JinaBertConfig, JinaBertModel
+
+    >>> # Initializing a JinaBert configuration
+    >>> configuration = JinaBertConfig()
+
+    >>> # Initializing a model (with random weights) from the configuration
+    >>> model = JinaBertModel(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+
+    >>> # Encode text inputs
+    >>> embeddings = model.encode(text_inputs)
+    ```"""
+    model_type = "bert"
+
+    def __init__(
+        self,
+        vocab_size=30522,
+        hidden_size=768,
+        num_hidden_layers=12,
+        num_attention_heads=12,
+        intermediate_size=3072,
+        hidden_act="gelu",
+        hidden_dropout_prob=0.1,
+        attention_probs_dropout_prob=0.1,
+        max_position_embeddings=512,
+        type_vocab_size=2,
+        initializer_range=0.02,
+        layer_norm_eps=1e-12,
+        pad_token_id=0,
+        position_embedding_type="absolute",
+        use_cache=True,
+        classifier_dropout=None,
+        feed_forward_type="original",
+        emb_pooler=None,
+        attn_implementation='torch',
+        **kwargs,
+    ):
+        super().__init__(pad_token_id=pad_token_id, **kwargs)
+
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.hidden_act = hidden_act
+        self.intermediate_size = intermediate_size
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.max_position_embeddings = max_position_embeddings
+        self.type_vocab_size = type_vocab_size
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
+        self.position_embedding_type = position_embedding_type
+        self.use_cache = use_cache
+        self.classifier_dropout = classifier_dropout
+        self.feed_forward_type = feed_forward_type
+        self.emb_pooler = emb_pooler
+        self.attn_implementation = attn_implementation
+
+class JinaBertOnnxConfig(OnnxConfig):
+    @property
+    def inputs(self) -> Mapping[str, Mapping[int, str]]:
+        if self.task == "multiple-choice":
+            dynamic_axis = {0: "batch", 1: "choice", 2: "sequence"}
+        else:
+            dynamic_axis = {0: "batch", 1: "sequence"}
+        return OrderedDict(
+            [
+                ("input_ids", dynamic_axis),
+                ("attention_mask", dynamic_axis),
+                ("token_type_ids", dynamic_axis),
+            ]
+        )

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7e9ea4e0b0879e9624fd0606f02b85384fe209ce5bc7cf5daecaf7e3fecf82f
+size 66100274

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77bbb3421aa3dca1886e8adcd0731bc1ca529a233266a4183278e43dffcaced8
+size 66106938

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}