Add SetFit model

README.md

@@ -5,65 +5,45 @@ tags:
 - sentence-transformers
 - text-classification
 pipeline_tag: text-classification
-language:
-- en
-metrics:
-- accuracy
 ---
 
 # Kaleemullah/paraphrase-mpnet-base-v2-ads-nonads-classifier
 
-## Introduction
-The "paraphrase-mpnet-base-v2-ads-nonads-classifier" is a SetFit model tailored for efficient text classification, specifically designed to differentiate between ad-related and non-ad content. This model leverages advanced few-shot learning techniques, making it a powerful tool for applications in content analysis and digital marketing.
+This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
 
-## Model Description
-This model is built upon a Sentence Transformer, fine-tuned through contrastive learning for enhanced feature extraction capabilities. It incorporates a classification head trained with these fine-tuned features, enabling high accuracy and efficiency in classifying texts into ad-related or non-ad categories. The model is particularly effective in processing and understanding the subtle differences in language used in advertising content.
+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.
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/61b2c130ac5ecaae3d1efe27/zS37gyEQW7MhP9CwWb7zZ.png)
+## Usage
 
-## Training Data
-The model was trained on a comprehensive dataset consisting of various examples of ad-related and non-ad content sourced from multiple digital marketing platforms. The dataset includes a diverse range of linguistic styles and terminologies, ensuring robustness and versatility in the model's classification ability.
-
-## Performance Metrics
-The model performs excellently in text classification tasks, highlighted by its accuracy, precision, recall, and F1 score metrics. These metrics underscore the model's effectiveness in correctly identifying and classifying ad and non-ad content.
-
-## Installation and Usage
-
-To install the SetFit library, use one of the following commands:
+To use this model for inference, first install the SetFit library:
 
 ```bash
-python3 -m pip install setfit
-# or
-pip install setfit
+python -m pip install setfit
 ```
 
-
-You can then run the inference as follows:
+You can then run inference as follows:
 
 ```python
-
 from setfit import SetFitModel
 
-# Initialize and download the model from the HuggingFace Hub
+# Download from Hub and run inference
 model = SetFitModel.from_pretrained("Kaleemullah/paraphrase-mpnet-base-v2-ads-nonads-classifier")
-
-# Example texts for classification
-texts = ["xiaomi phone best with best camera quality ", "Pineapple on pizza is the worst 🤮"]
-
 # Run inference
-# This will classify each text in the 'texts' list as ad-related or non-ad content
-predictions = model(texts)
-
-
-# 'predictions' will be a list of labels corresponding to each input text
-for text, pred in zip(texts, predictions):
-    print(f"Text: {text}\nPredicted Category: {'Ad' if pred else 'Non-Ad'}\n")
+preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"])
 ```
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/61b2c130ac5ecaae3d1efe27/63Flu2lnmR6XS6E2hm9MY.png)
-
-## Some Other Non-Ads Content Example 
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/61b2c130ac5ecaae3d1efe27/rZzbGFWXb4tYRhcO46LB7.png)
-
-
+## BibTeX entry and citation info
+
+```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}
+}
+```

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/root/.cache/torch/sentence_transformers/sentence-transformers_paraphrase-mpnet-base-v2/",
+  "_name_or_path": "/root/.cache/torch/sentence_transformers/Kaleemullah_paraphrase-mpnet-base-v2-ads-nonads-classifier/",
   "architectures": [
     "MPNetModel"
   ],

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3f1f75a5936d708c2e06911e77e15b1908bf8e75650ce8b3bb36c4c0cf53810
+oid sha256:0faec4a3c5dd1650ceeb4068d6b6c862833cd0bc26904c900ca9bc057e7d5fbb
 size 437967672

model_head.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b8adb7b03dd706b76fe6f8fc5776b9b1e00e85954abe4d0a75bc61989efd030
+oid sha256:ce7c32d373f1de1eb51580dadc989970716a38e1e7c975fb683af0f456aaafd0
 size 7706

special_tokens_map.json

@@ -9,7 +9,7 @@
   "cls_token": {
     "content": "<s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
@@ -37,7 +37,7 @@
   "sep_token": {
     "content": "</s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },

tokenizer_config.json

@@ -48,12 +48,19 @@
   "do_lower_case": true,
   "eos_token": "</s>",
   "mask_token": "<mask>",
+  "max_length": 512,
   "model_max_length": 512,
   "never_split": null,
+  "pad_to_multiple_of": null,
   "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
   "sep_token": "</s>",
+  "stride": 0,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "MPNetTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
   "unk_token": "[UNK]"
 }