Push model using huggingface_hub.

1_Pooling/config.json

@@ -1,10 +1,5 @@
 {
-    "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,
+    "embedding_dimension": 384,
+    "pooling_mode": "mean",
     "include_prompt": true
 }

README.md

@@ -3,281 +3,93 @@ tags:
 - setfit
 - sentence-transformers
 - text-classification
-- code-security
-- dlp
-- secret-detection
-- vulnerability-detection
-language:
-- en
-license: apache-2.0
-widget:
-- text: "Analyze this hardcoded_secret: AWS_ACCESS_KEY_ID=AKIA4REALKEY123ABC committed to main branch .env file"
-- text: "Analyze this vulnerable_pattern: $query = 'SELECT * FROM users WHERE id=' . $_GET['id'];"
-- text: "Analyze this test_fixture: factory_boy default: user.password = 'testpass123' for pytest fixtures"
-- text: "Analyze this clean_code: def get_user(db: Session, user_id: int): return db.query(User).filter(User.id == user_id).first()"
-- text: "Analyze this secure_implementation: password_hash = bcrypt.hashpw(password.encode(), bcrypt.gensalt(rounds=12))"
+- generated_from_setfit_trainer
+widget: []
 metrics:
 - accuracy
 pipeline_tag: text-classification
 library_name: setfit
 inference: true
-model-index:
-- name: spidercob/code-risk-classifier
-  results:
-  - task:
-      type: text-classification
-      name: Text Classification
-    dataset:
-      name: curated-public-repos-v2
-      type: custom
-      split: test
-    metrics:
-    - type: accuracy
-      value: 1.0
-      name: Accuracy
 ---
 
-# spidercob/code-risk-classifier
+# SetFit
 
-A [SetFit](https://github.com/huggingface/setfit) model that classifies code snippets and secrets into four risk categories. Used inside [Spidercob](https://spidercob.com) to reduce false positives in supply-chain and secret scanning — distinguishing real production risks from test fixtures and safe code.
+This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
 
-**v2** — expanded multi-language training corpus (22 public repos, Python / Java / PHP / JavaScript / TypeScript / Ruby / Go).
+The model has been trained using an efficient few-shot learning technique that involves:
 
-## Labels
+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.
 
-| Label | Meaning | Action |
-|---|---|---|
-| `REAL_SECRET` | Hardcoded credential, API key, or token committed to source | **BLOCK** |
-| `VULNERABLE_LOGIC` | SQL injection, XSS, unsafe deserialization, command injection, etc. | **BLOCK** |
-| `TEST_MOCK` | Dummy credential or vuln pattern inside a test fixture or factory | ALLOW |
-| `SAFE_CODE` | Clean production code, secure implementation pattern | ALLOW |
+## Model Details
 
-## Quick Start
+### Model Description
+- **Model Type:** SetFit
+<!-- - **Sentence Transformer:** [Unknown](https://huggingface.co/unknown) -->
+- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
+- **Maximum Sequence Length:** 256 tokens
+- **Number of Classes:** 4 classes
+<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
 
-```python
-from setfit import SetFitModel
+### 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)
 
-model = SetFitModel.from_pretrained("spidercob/code-risk-classifier")
+## Uses
 
-snippets = [
-    "Analyze this hardcoded_secret: AWS_ACCESS_KEY_ID=AKIA4REALKEY123ABC in .env",
-    "Analyze this vulnerable_pattern: $q = 'SELECT * FROM users WHERE id=' . $_GET['id'];",
-    "Analyze this test_fixture: user.password = 'testpass123'  # factory_boy default",
-    "Analyze this clean_code: password_hash = bcrypt.hashpw(password.encode(), bcrypt.gensalt())",
-]
+### Direct Use for Inference
 
-predictions = model.predict(snippets)
-probabilities = model.predict_proba(snippets)
-# predictions: ['REAL_SECRET', 'VULNERABLE_LOGIC', 'TEST_MOCK', 'SAFE_CODE']
+First install the SetFit library:
+
+```bash
+pip install setfit
 ```
 
-## Input Format
+Then you can load this model and run inference.
 
-Inputs must follow the pattern used during training:
+```python
+from setfit import SetFitModel
 
-```
-Analyze this <issue_type>: <code_snippet_or_context>
+# Download from the 🤗 Hub
+model = SetFitModel.from_pretrained("setfit_model_id")
+# Run inference
+preds = model("I loved the spiderman movie!")
 ```
 
-Where `<issue_type>` is a short descriptor such as `hardcoded_secret`, `vulnerable_pattern`, `test_fixture`, `clean_code`, `secure_implementation`, etc.
+<!--
+### Downstream Use
 
-## Integration with Spidercob DLP
+*List how someone could finetune this model on their own dataset.*
+-->
 
-```python
-from app.core.code_risk_classifier import CodeRiskClassifier
+<!--
+### Out-of-Scope Use
 
-clf = CodeRiskClassifier()
-result = clf.classify("hardcoded_secret", "STRIPE_SECRET_KEY = 'sk_live_abc123xyz'")
-# result: {"risk_type": "REAL_SECRET", "confidence": 0.99, "action": "BLOCK", "severity": "CRITICAL"}
-```
+*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.*
+-->
 
-The classifier loads the fine-tuned SetFit model at import time and falls back to zero-shot cosine similarity if the model directory is absent.
-
-## Evaluation
-
-| Label   | Accuracy |
-|:--------|:---------|
-| **all** | **1.0** (100%, 176 test examples) |
-
-## Training Details (v2)
-
-### Dataset
-
-877 labelled examples extracted from **22 curated public GitHub repositories** via regex-based heuristics. Each example is a 3–7 line context window around a matched line, formatted as `Analyze this <issue_type>: <context>`.
-
-**Languages covered:** Python, Java, PHP, JavaScript, TypeScript, Ruby, Go
-
-### Training Set Composition
-
-| Label | Train examples | Source strategy |
-|---|---|---|
-| `REAL_SECRET` | 27 | TruffleHog test corpus (confirmed leaked secrets), Railsgoat, NodeGoat, Juice Shop |
-| `VULNERABLE_LOGIC` | 194 | DVWA (PHP), WebGoat (Java), vulhub, NodeGoat, DVGA, Railsgoat, Juice Shop |
-| `TEST_MOCK` | 240 | factory_boy, Faker, pytest, model_bakery |
-| `SAFE_CODE` | 240 | Django, FastAPI, requests, Flask, httpx, Devise, Sinatra, Gin |
-
-### Source Repositories
-
-| Repo | Label | Language |
-|---|---|---|
-| OWASP/WebGoat | VULNERABLE_LOGIC | Java |
-| digininja/DVWA | VULNERABLE_LOGIC | PHP |
-| vulhub/vulhub | VULNERABLE_LOGIC | multi |
-| trufflesecurity/trufflehog | REAL_SECRET | Go/multi |
-| OWASP/NodeGoat | VULNERABLE_LOGIC | JavaScript |
-| dolevf/Damn-Vulnerable-GraphQL-Application | VULNERABLE_LOGIC | Python |
-| OWASP/railsgoat | VULNERABLE_LOGIC | Ruby |
-| juice-shop/juice-shop | VULNERABLE_LOGIC | TypeScript |
-| FactoryBoy/factory_boy | TEST_MOCK | Python |
-| joke2k/faker | TEST_MOCK | Python |
-| pytest-dev/pytest | TEST_MOCK | Python |
-| model-bakers/model_bakery | TEST_MOCK | Python |
-| django/django | SAFE_CODE | Python |
-| tiangolo/fastapi | SAFE_CODE | Python |
-| psf/requests | SAFE_CODE | Python |
-| pallets/flask | SAFE_CODE | Python |
-| encode/httpx | SAFE_CODE | Python |
-| heartcombo/devise | SAFE_CODE | Ruby |
-| sinatra/sinatra | SAFE_CODE | Ruby |
-| gin-gonic/gin | SAFE_CODE | Go |
-| golang/vulndb | VULNERABLE_LOGIC | Go |
-
-### Training Hyperparameters
-- batch_size: (16, 16)
-- num_epochs: (3, 3)
-- max_steps: -1
-- sampling_strategy: oversampling
-- num_iterations: 20
-- body_learning_rate: (2e-05, 1e-05)
-- head_learning_rate: 0.01
-- loss: CosineSimilarityLoss
-- distance_metric: cosine_distance
-- margin: 0.25
-- end_to_end: False
-- use_amp: False
-- warmup_proportion: 0.1
-- l2_weight: 0.01
-- seed: 42
-- eval_max_steps: -1
-- load_best_model_at_end: True
-
-### Training Results
-| Epoch  | Step | Training Loss | Validation Loss |
-|:------:|:----:|:-------------:|:---------------:|
-| 0.0006 | 1    | 0.0416        | -               |
-| 0.0285 | 50   | 0.0099        | -               |
-| 0.0570 | 100  | 0.0028        | -               |
-| 0.0856 | 150  | 0.001         | -               |
-| 0.1141 | 200  | 0.0013        | -               |
-| 0.1426 | 250  | 0.0003        | -               |
-| 0.1711 | 300  | 0.0002        | -               |
-| 0.1997 | 350  | 0.0002        | -               |
-| 0.2282 | 400  | 0.0002        | -               |
-| 0.2567 | 450  | 0.0002        | -               |
-| 0.2852 | 500  | 0.0002        | -               |
-| 0.3137 | 550  | 0.0001        | -               |
-| 0.3423 | 600  | 0.0001        | -               |
-| 0.3708 | 650  | 0.0001        | -               |
-| 0.3993 | 700  | 0.0001        | -               |
-| 0.4278 | 750  | 0.0001        | -               |
-| 0.4564 | 800  | 0.0001        | -               |
-| 0.4849 | 850  | 0.0001        | -               |
-| 0.5134 | 900  | 0.0001        | -               |
-| 0.5419 | 950  | 0.0001        | -               |
-| 0.5705 | 1000 | 0.0001        | -               |
-| 0.5990 | 1050 | 0.0001        | -               |
-| 0.6275 | 1100 | 0.0001        | -               |
-| 0.6560 | 1150 | 0.0001        | -               |
-| 0.6845 | 1200 | 0.0001        | -               |
-| 0.7131 | 1250 | 0.0001        | -               |
-| 0.7416 | 1300 | 0.0001        | -               |
-| 0.7701 | 1350 | 0.0001        | -               |
-| 0.7986 | 1400 | 0.0           | -               |
-| 0.8272 | 1450 | 0.0001        | -               |
-| 0.8557 | 1500 | 0.0001        | -               |
-| 0.8842 | 1550 | 0.0           | -               |
-| 0.9127 | 1600 | 0.0           | -               |
-| 0.9412 | 1650 | 0.0001        | -               |
-| 0.9698 | 1700 | 0.0           | -               |
-| 0.9983 | 1750 | 0.0           | -               |
-| 1.0    | 1753 | -             | 0.0001          |
-| 1.0268 | 1800 | 0.0           | -               |
-| 1.0553 | 1850 | 0.0           | -               |
-| 1.0839 | 1900 | 0.0           | -               |
-| 1.1124 | 1950 | 0.0           | -               |
-| 1.1409 | 2000 | 0.0           | -               |
-| 1.1694 | 2050 | 0.0           | -               |
-| 1.1979 | 2100 | 0.0           | -               |
-| 1.2265 | 2150 | 0.0           | -               |
-| 1.2550 | 2200 | 0.0           | -               |
-| 1.2835 | 2250 | 0.0           | -               |
-| 1.3120 | 2300 | 0.0           | -               |
-| 1.3406 | 2350 | 0.0           | -               |
-| 1.3691 | 2400 | 0.0           | -               |
-| 1.3976 | 2450 | 0.0           | -               |
-| 1.4261 | 2500 | 0.0           | -               |
-| 1.4546 | 2550 | 0.0           | -               |
-| 1.4832 | 2600 | 0.0           | -               |
-| 1.5117 | 2650 | 0.0           | -               |
-| 1.5402 | 2700 | 0.0           | -               |
-| 1.5687 | 2750 | 0.0           | -               |
-| 1.5973 | 2800 | 0.0           | -               |
-| 1.6258 | 2850 | 0.0           | -               |
-| 1.6543 | 2900 | 0.0           | -               |
-| 1.6828 | 2950 | 0.0           | -               |
-| 1.7114 | 3000 | 0.0           | -               |
-| 1.7399 | 3050 | 0.0           | -               |
-| 1.7684 | 3100 | 0.0           | -               |
-| 1.7969 | 3150 | 0.0           | -               |
-| 1.8254 | 3200 | 0.0           | -               |
-| 1.8540 | 3250 | 0.0           | -               |
-| 1.8825 | 3300 | 0.0           | -               |
-| 1.9110 | 3350 | 0.0           | -               |
-| 1.9395 | 3400 | 0.0           | -               |
-| 1.9681 | 3450 | 0.0           | -               |
-| 1.9966 | 3500 | 0.0           | -               |
-| 2.0    | 3506 | -             | 0.0000          |
-| 2.0251 | 3550 | 0.0           | -               |
-| 2.0536 | 3600 | 0.0           | -               |
-| 2.0821 | 3650 | 0.0           | -               |
-| 2.1107 | 3700 | 0.0           | -               |
-| 2.1392 | 3750 | 0.0           | -               |
-| 2.1677 | 3800 | 0.0           | -               |
-| 2.1962 | 3850 | 0.0           | -               |
-| 2.2248 | 3900 | 0.0           | -               |
-| 2.2533 | 3950 | 0.0           | -               |
-| 2.2818 | 4000 | 0.0           | -               |
-| 2.3103 | 4050 | 0.0           | -               |
-| 2.3388 | 4100 | 0.0           | -               |
-| 2.3674 | 4150 | 0.0           | -               |
-| 2.3959 | 4200 | 0.0           | -               |
-| 2.4244 | 4250 | 0.0           | -               |
-| 2.4529 | 4300 | 0.0           | -               |
-| 2.4815 | 4350 | 0.0           | -               |
-| 2.5100 | 4400 | 0.0           | -               |
-| 2.5385 | 4450 | 0.0           | -               |
-| 2.5670 | 4500 | 0.0           | -               |
-| 2.5956 | 4550 | 0.0           | -               |
-| 2.6241 | 4600 | 0.0           | -               |
-| 2.6526 | 4650 | 0.0           | -               |
-| 2.6811 | 4700 | 0.0           | -               |
-| 2.7096 | 4750 | 0.0           | -               |
-| 2.7382 | 4800 | 0.0           | -               |
-| 2.7667 | 4850 | 0.0           | -               |
-| 2.7952 | 4900 | 0.0           | -               |
-| 2.8237 | 4950 | 0.0           | -               |
-| 2.8523 | 5000 | 0.0           | -               |
-| 2.8808 | 5050 | 0.0           | -               |
-| 2.9093 | 5100 | 0.0           | -               |
-| 2.9378 | 5150 | 0.0           | -               |
-| 2.9663 | 5200 | 0.0           | -               |
-| 2.9949 | 5250 | 0.0           | -               |
-| 3.0    | 5259 | -             | 0.0000          |
+## Training Details
 
 ### Framework Versions
 - Python: 3.12.12
 - SetFit: 1.1.3
-- Sentence Transformers: 5.2.2
+- Sentence Transformers: 5.6.1
 - Transformers: 4.57.6
 - PyTorch: 2.10.0
 - Datasets: 5.0.0

config_sentence_transformers.json

@@ -1,14 +1,14 @@
 {
   "__version__": {
-    "sentence_transformers": "5.2.2",
-    "transformers": "4.57.6",
-    "pytorch": "2.10.0"
+    "pytorch": "2.10.0",
+    "sentence_transformers": "5.6.1",
+    "transformers": "4.57.6"
   },
+  "default_prompt_name": null,
   "model_type": "SentenceTransformer",
   "prompts": {
-    "query": "",
-    "document": ""
+    "document": "",
+    "query": ""
   },
-  "default_prompt_name": null,
   "similarity_fn_name": "cosine"
 }

config_setfit.json

@@ -1,9 +1,9 @@
 {
-  "normalize_embeddings": false,
   "labels": [
     "REAL_SECRET",
     "VULNERABLE_LOGIC",
     "TEST_MOCK",
     "SAFE_CODE"
-  ]
+  ],
+  "normalize_embeddings": false
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c4c93bb1d3444cdc0eec9908b2e6e61660c7606b020a6776b3bd450f315bee7
+oid sha256:3f5e69a66f4af290805d351cbfd5968f48bb82dacf20a059800ca82265db9f6c
 size 90864192

model_head.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e9f45b9e642972266844fc03150c434167110c84d54306ba7119d5b0fea50fc
+oid sha256:908ee4643be4f4e547ae40fc8efe3dbf933f857655e62a9864a83e5389ea22e8
 size 13191

modules.json

@@ -3,18 +3,18 @@
     "idx": 0,
     "name": "0",
     "path": "",
-    "type": "sentence_transformers.models.Transformer"
+    "type": "sentence_transformers.base.modules.transformer.Transformer"
   },
   {
     "idx": 1,
     "name": "1",
     "path": "1_Pooling",
-    "type": "sentence_transformers.models.Pooling"
+    "type": "sentence_transformers.sentence_transformer.modules.pooling.Pooling"
   },
   {
     "idx": 2,
     "name": "2",
     "path": "2_Normalize",
-    "type": "sentence_transformers.models.Normalize"
+    "type": "sentence_transformers.sentence_transformer.modules.normalize.Normalize"
   }
 ]

sentence_bert_config.json

@@ -1,4 +1,10 @@
 {
-    "max_seq_length": 256,
-    "do_lower_case": false
+    "transformer_task": "feature-extraction",
+    "modality_config": {
+        "text": {
+            "method": "forward",
+            "method_output_name": "last_hidden_state"
+        }
+    },
+    "module_output_name": "token_embeddings"
 }