Upload 14 files

Add fine-tuned Model

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 384,
+    "pooling_mode_cls_token": false,
+    "pooling_mode_mean_tokens": true,
+    "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

@@ -1,3 +1,164 @@
----
-license: mit
----
+# SetFit Email Job Status Classifier
+
+A fine-tuned SetFit model for classifying job application emails into different categories. This model can automatically categorize job-related emails to help track application status and organize your job search process.
+
+## Model Description
+
+This model is based on [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) and fine-tuned using SetFit (Sentence Transformer Fine-tuning) for email classification tasks.
+
+**Base Model:** `sentence-transformers/all-MiniLM-L6-v2`  
+**Framework:** SetFit  
+**Task:** Multi-class text classification  
+**Language:** English
+
+## Model Performance
+
+- **Overall Accuracy:** 90%
+- **Training Samples:** 235
+- **Test Samples:** 59
+- **Classes:** 7
+
+## Intended Use
+
+### Primary Use Cases
+
+- **Job Search Management:** Automatically categorize job-related emails
+- **Email Organization:** Sort incoming emails by job application status
+- **Application Tracking:** Monitor progress of job applications
+- **Workflow Automation:** Integrate into email filtering systems
+
+### Supported Categories
+
+1. **Applied** - Confirmation emails after submitting applications
+2. **Interview** - Interview invitations and scheduling emails
+3. **Next-Step** - Follow-up emails about next steps in the process
+4. **Not Job Related** - Non-job related emails
+5. **Not Job Status Update** - Job-related but not status updates
+6. **Offer** - Job offers and offer-related communications
+7. **Rejected** - Rejection notifications
+
+## How to Use
+
+### Installation
+
+```bash
+pip install setfit
+```
+
+### Quick Start
+
+```python
+from setfit import SetFitModel
+
+# Load the model
+model = SetFitModel.from_pretrained("your-username/setfit-email-classifier")
+
+# Classify emails
+emails = [
+    "Thank you for your application. We will review it and get back to you.",
+    "We would like to schedule an interview with you for next Tuesday.",
+    "Unfortunately, we have decided to move forward with other candidates.",
+    "Congratulations! We would like to offer you the position."
+]
+
+predictions = model(emails)
+print(predictions)
+# Output: ['applied', 'interview', 'rejected', 'offer']
+```
+
+### Batch Processing
+
+```python
+import pandas as pd
+
+# Process multiple emails
+df = pd.DataFrame({
+    'subject': ['Application Confirmation', 'Interview Invitation', 'Unfortunately...'],
+    'body': ['Thank you for applying...', 'We would like to schedule...', 'We regret to inform...']
+})
+
+# Combine subject and body
+df['email_text'] = df['subject'] + ' ' + df['body']
+
+# Get predictions
+predictions = model(df['email_text'].tolist())
+df['classification'] = predictions
+
+print(df[['subject', 'classification']])
+```
+
+### Confidence Scores
+
+```python
+# Get prediction probabilities
+email_text = "We would like to schedule a phone interview with you."
+embeddings = model.model_body.encode([email_text])
+probabilities = model.model_head.predict_proba(embeddings)[0]
+classes = model.model_head.classes_
+
+# Display results
+prediction = model([email_text])[0]
+print(f"Prediction: {prediction}")
+
+for class_name, prob in zip(classes, probabilities):
+    print(f"{class_name}: {prob:.3f}")
+```
+
+## Training Data
+
+The model was trained on a dataset of 294 job-related emails with the following distribution:
+
+- **Not Job Related:** 53 samples
+- **Rejected:** 47 samples
+- **Interview:** 47 samples
+- **Not Job Status Update:** 46 samples
+- **Applied:** 43 samples
+- **Next-Step:** 30 samples
+- **Offer:** 28 samples
+
+## Training Details
+
+**Training Framework:** SetFit  
+**Training Time:** ~1 hour  
+**Hardware:** Local training (CPU/GPU)  
+**Optimization:** Contrastive learning + classification head fine-tuning
+
+### Training Parameters
+
+- **Base Model:** sentence-transformers/all-MiniLM-L6-v2
+- **Train/Test Split:** 80/20
+- **Stratified Split:** Yes (maintains class distribution)
+- **Random State:** 42 (reproducible results)
+
+## Limitations
+
+- **Domain Specific:** Trained specifically on job application emails
+- **English Only:** Optimized for English language emails
+- **Class Imbalance:** Some categories have fewer training examples
+- **Context Length:** Best performance on email-length texts (not very long documents)
+- **Temporal Drift:** Performance may degrade on very recent email formats/styles
+
+## Bias and Fairness
+
+- The model may reflect biases present in the training data
+- Performance may vary across different industries, company sizes, or communication styles
+- Regular evaluation recommended when deployed in production
+
+## Citation
+
+If you use this model, please cite:
+
+```bibtex
+@misc{setfit-email-classifier-2025,
+  title={SetFit Email Job Status Classifier},
+  author={Oluwatomiwa Jinadu},
+  year={2025},
+  howpublished={\\url{https://huggingface.co/Tomiwajin/setfit_email_classifier}},
+}
+```
+
+---
+
+**Model Card Authors:** Oluwatomiwa Jinadu
+**Model Card Date:** September 2025  
+**Model Version:** 1.0

config.json

@@ -0,0 +1,25 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "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": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.56.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "sentence_transformers": "5.1.0",
+    "transformers": "4.56.1",
+    "pytorch": "2.2.2"
+  },
+  "model_type": "SentenceTransformer",
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

config_setfit.json

@@ -0,0 +1,12 @@
+{
+  "labels": [
+    "applied",
+    "interview",
+    "next-step",
+    "not_job_related",
+    "not_job_status_update",
+    "offer",
+    "rejected"
+  ],
+  "normalize_embeddings": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:deec023621bfb0a09323c23fcb6a2b2cd73367b983cd8ae61425e23d91084a0e
+size 90864192

model_head.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4203f0bf3387d026cb71eb9b2e4fa5262c460b42e7052d57d10296112c7f1a5a
+size 22991

model_info.json

@@ -0,0 +1,18 @@
+{
+  "model_path": "setfit_email_classifier_20250909_135238",
+  "timestamp": "20250909_135238",
+  "base_model": "sentence-transformers/all-MiniLM-L6-v2",
+  "training_date": "2025-09-09T13:52:39.181047",
+  "model_type": "SetFit Email Classifier",
+  "training_samples": 235,
+  "test_samples": 59,
+  "classes": [
+    "applied",
+    "interview",
+    "next-step",
+    "not_job_related",
+    "not_job_status_update",
+    "offer",
+    "rejected"
+  ]
+}

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": 256,
+    "do_lower_case": false
+}

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 +1,65 @@
+{
+  "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": false,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "max_length": 128,
+  "model_max_length": 256,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}