Upload MLX packaging for OpenMed-PII-SuperMedical-Base-125M-v1-mlx

README.md

@@ -0,0 +1,105 @@
+---
+license: apache-2.0
+base_model: OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1
+pipeline_tag: token-classification
+library_name: openmed
+tags:
+  - openmed
+  - mlx
+  - apple-silicon
+  - token-classification
+  - pii
+  - de-identification
+  - medical
+  - clinical
+---
+
+# OpenMed-PII-SuperMedical-Base-125M-v1 for OpenMed MLX
+
+This repository contains a **private MLX packaging** of [`OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1`](https://huggingface.co/OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1) for Apple Silicon inference with [**OpenMed**](https://github.com/maziyarpanahi/openmed).
+
+[OpenMed](https://github.com/maziyarpanahi/openmed) is the main product experience:
+
+- Install the Python package with `pip install openmed`
+- Enable Apple Silicon acceleration with `pip install "openmed[mlx]"`
+- Run the same OpenMed API you already use, now backed by MLX on macOS
+- For Apple apps, use **OpenMedKit** from the same GitHub repository with a CoreML bundle
+
+This MLX repo is meant to pair with:
+
+- OpenMed GitHub: [https://github.com/maziyarpanahi/openmed](https://github.com/maziyarpanahi/openmed)
+- OpenMed website: [https://openmed.life](https://openmed.life)
+- Source checkpoint: [`OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1`](https://huggingface.co/OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1)
+
+## Quick Start
+
+### Python
+
+```bash
+pip install openmed
+pip install "openmed[mlx]"
+```
+
+```python
+from openmed import analyze_text
+from openmed.core.config import OpenMedConfig
+
+result = analyze_text(
+    "Patient John Doe, DOB 1990-05-15, SSN 123-45-6789",
+    model_name="OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1",
+    config=OpenMedConfig(backend="mlx"),
+)
+
+for entity in result.entities:
+    print(entity.label, entity.text, round(entity.confidence, 4))
+```
+
+You can use the source model ID directly through OpenMed, or download this private MLX packaging from [`OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1-mlx`](https://huggingface.co/OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1-mlx) for a preconverted path.
+
+### Swift
+
+Use Swift with **OpenMedKit**, not with MLX weight files directly.
+
+1. Open Xcode and go to **File > Add Package Dependencies...**
+2. Paste the OpenMed repository URL:
+   `https://github.com/maziyarpanahi/openmed`
+3. Choose the package product **OpenMedKit** from the repository.
+4. Add a compatible CoreML model bundle plus `id2label.json` to your app target.
+
+After that, import OpenMedKit in Swift:
+
+```swift
+import OpenMedKit
+```
+
+Then load your bundled CoreML model and label map:
+
+```swift
+import Foundation
+import OpenMedKit
+
+let modelFolder = Bundle.main.resourceURL!
+let modelURL = modelFolder.appendingPathComponent("OpenMedPII.mlmodelc")
+let labelsURL = modelFolder.appendingPathComponent("id2label.json")
+
+let openmed = try OpenMed(
+    modelURL: modelURL,
+    id2labelURL: labelsURL
+)
+```
+
+This private MLX artifact remains the right choice for:
+
+- Python services on Apple Silicon
+- local MLX inference on macOS
+- private preconverted packaging on the Hub
+
+If a given architecture or environment cannot be exported cleanly as `safetensors`, OpenMed falls back to `weights.npz` so the model remains usable.
+
+## Credits
+
+- Base checkpoint: [`OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1`](https://huggingface.co/OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1)
+- OpenMed GitHub: [https://github.com/maziyarpanahi/openmed](https://github.com/maziyarpanahi/openmed)
+- OpenMed website: [https://openmed.life](https://openmed.life)
+- MLX packaging and runtime support: [**OpenMed**](https://github.com/maziyarpanahi/openmed)
+- Swift runtime for Apple apps: **OpenMedKit** from the OpenMed repository

config.json

@@ -0,0 +1,256 @@
+{
+  "transformers_version": "5.5.0",
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "output_hidden_states": false,
+  "return_dict": true,
+  "dtype": "float32",
+  "chunk_size_feed_forward": 0,
+  "is_encoder_decoder": false,
+  "id2label": {
+    "0": "O",
+    "1": "B-account_number",
+    "2": "B-age",
+    "3": "B-api_key",
+    "4": "B-bank_routing_number",
+    "5": "B-biometric_identifier",
+    "6": "B-blood_type",
+    "7": "B-certificate_license_number",
+    "8": "B-city",
+    "9": "B-company_name",
+    "10": "B-coordinate",
+    "11": "B-country",
+    "12": "B-county",
+    "13": "B-credit_debit_card",
+    "14": "B-customer_id",
+    "15": "B-cvv",
+    "16": "B-date",
+    "17": "B-date_of_birth",
+    "18": "B-date_time",
+    "19": "B-device_identifier",
+    "20": "B-education_level",
+    "21": "B-email",
+    "22": "B-employee_id",
+    "23": "B-employment_status",
+    "24": "B-fax_number",
+    "25": "B-first_name",
+    "26": "B-gender",
+    "27": "B-health_plan_beneficiary_number",
+    "28": "B-http_cookie",
+    "29": "B-ipv4",
+    "30": "B-ipv6",
+    "31": "B-language",
+    "32": "B-last_name",
+    "33": "B-license_plate",
+    "34": "B-mac_address",
+    "35": "B-medical_record_number",
+    "36": "B-occupation",
+    "37": "B-password",
+    "38": "B-phone_number",
+    "39": "B-pin",
+    "40": "B-political_view",
+    "41": "B-postcode",
+    "42": "B-race_ethnicity",
+    "43": "B-religious_belief",
+    "44": "B-sexuality",
+    "45": "B-ssn",
+    "46": "B-state",
+    "47": "B-street_address",
+    "48": "B-swift_bic",
+    "49": "B-tax_id",
+    "50": "B-time",
+    "51": "B-unique_id",
+    "52": "B-url",
+    "53": "B-user_name",
+    "54": "B-vehicle_identifier",
+    "55": "I-account_number",
+    "56": "I-api_key",
+    "57": "I-biometric_identifier",
+    "58": "I-blood_type",
+    "59": "I-certificate_license_number",
+    "60": "I-city",
+    "61": "I-company_name",
+    "62": "I-coordinate",
+    "63": "I-country",
+    "64": "I-county",
+    "65": "I-credit_debit_card",
+    "66": "I-customer_id",
+    "67": "I-date",
+    "68": "I-date_of_birth",
+    "69": "I-date_time",
+    "70": "I-device_identifier",
+    "71": "I-education_level",
+    "72": "I-email",
+    "73": "I-employee_id",
+    "74": "I-employment_status",
+    "75": "I-fax_number",
+    "76": "I-first_name",
+    "77": "I-gender",
+    "78": "I-health_plan_beneficiary_number",
+    "79": "I-http_cookie",
+    "80": "I-ipv4",
+    "81": "I-ipv6",
+    "82": "I-language",
+    "83": "I-last_name",
+    "84": "I-license_plate",
+    "85": "I-mac_address",
+    "86": "I-medical_record_number",
+    "87": "I-occupation",
+    "88": "I-password",
+    "89": "I-phone_number",
+    "90": "I-pin",
+    "91": "I-political_view",
+    "92": "I-postcode",
+    "93": "I-race_ethnicity",
+    "94": "I-religious_belief",
+    "95": "I-sexuality",
+    "96": "I-ssn",
+    "97": "I-state",
+    "98": "I-street_address",
+    "99": "I-swift_bic",
+    "100": "I-tax_id",
+    "101": "I-time",
+    "102": "I-unique_id",
+    "103": "I-url",
+    "104": "I-user_name",
+    "105": "I-vehicle_identifier"
+  },
+  "label2id": {
+    "B-account_number": 1,
+    "B-age": 2,
+    "B-api_key": 3,
+    "B-bank_routing_number": 4,
+    "B-biometric_identifier": 5,
+    "B-blood_type": 6,
+    "B-certificate_license_number": 7,
+    "B-city": 8,
+    "B-company_name": 9,
+    "B-coordinate": 10,
+    "B-country": 11,
+    "B-county": 12,
+    "B-credit_debit_card": 13,
+    "B-customer_id": 14,
+    "B-cvv": 15,
+    "B-date": 16,
+    "B-date_of_birth": 17,
+    "B-date_time": 18,
+    "B-device_identifier": 19,
+    "B-education_level": 20,
+    "B-email": 21,
+    "B-employee_id": 22,
+    "B-employment_status": 23,
+    "B-fax_number": 24,
+    "B-first_name": 25,
+    "B-gender": 26,
+    "B-health_plan_beneficiary_number": 27,
+    "B-http_cookie": 28,
+    "B-ipv4": 29,
+    "B-ipv6": 30,
+    "B-language": 31,
+    "B-last_name": 32,
+    "B-license_plate": 33,
+    "B-mac_address": 34,
+    "B-medical_record_number": 35,
+    "B-occupation": 36,
+    "B-password": 37,
+    "B-phone_number": 38,
+    "B-pin": 39,
+    "B-political_view": 40,
+    "B-postcode": 41,
+    "B-race_ethnicity": 42,
+    "B-religious_belief": 43,
+    "B-sexuality": 44,
+    "B-ssn": 45,
+    "B-state": 46,
+    "B-street_address": 47,
+    "B-swift_bic": 48,
+    "B-tax_id": 49,
+    "B-time": 50,
+    "B-unique_id": 51,
+    "B-url": 52,
+    "B-user_name": 53,
+    "B-vehicle_identifier": 54,
+    "I-account_number": 55,
+    "I-api_key": 56,
+    "I-biometric_identifier": 57,
+    "I-blood_type": 58,
+    "I-certificate_license_number": 59,
+    "I-city": 60,
+    "I-company_name": 61,
+    "I-coordinate": 62,
+    "I-country": 63,
+    "I-county": 64,
+    "I-credit_debit_card": 65,
+    "I-customer_id": 66,
+    "I-date": 67,
+    "I-date_of_birth": 68,
+    "I-date_time": 69,
+    "I-device_identifier": 70,
+    "I-education_level": 71,
+    "I-email": 72,
+    "I-employee_id": 73,
+    "I-employment_status": 74,
+    "I-fax_number": 75,
+    "I-first_name": 76,
+    "I-gender": 77,
+    "I-health_plan_beneficiary_number": 78,
+    "I-http_cookie": 79,
+    "I-ipv4": 80,
+    "I-ipv6": 81,
+    "I-language": 82,
+    "I-last_name": 83,
+    "I-license_plate": 84,
+    "I-mac_address": 85,
+    "I-medical_record_number": 86,
+    "I-occupation": 87,
+    "I-password": 88,
+    "I-phone_number": 89,
+    "I-pin": 90,
+    "I-political_view": 91,
+    "I-postcode": 92,
+    "I-race_ethnicity": 93,
+    "I-religious_belief": 94,
+    "I-sexuality": 95,
+    "I-ssn": 96,
+    "I-state": 97,
+    "I-street_address": 98,
+    "I-swift_bic": 99,
+    "I-tax_id": 100,
+    "I-time": 101,
+    "I-unique_id": 102,
+    "I-url": 103,
+    "I-user_name": 104,
+    "I-vehicle_identifier": 105,
+    "O": 0
+  },
+  "problem_type": null,
+  "vocab_size": 50265,
+  "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": 514,
+  "type_vocab_size": 1,
+  "initializer_range": 0.02,
+  "layer_norm_eps": 1e-05,
+  "pad_token_id": 1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "use_cache": true,
+  "classifier_dropout": null,
+  "is_decoder": false,
+  "add_cross_attention": false,
+  "tie_word_embeddings": true,
+  "_name_or_path": "OpenMed/OpenMed-PII-SuperMedical-Base-125M-v1",
+  "model_type": "roberta",
+  "position_embedding_type": "absolute",
+  "output_attentions": false,
+  "_mlx_position_offset": 2,
+  "_mlx_model_type": "bert",
+  "num_labels": 106,
+  "_mlx_weights_format": "safetensors"
+}

id2label.json

@@ -0,0 +1,108 @@
+{
+  "0": "O",
+  "1": "B-account_number",
+  "2": "B-age",
+  "3": "B-api_key",
+  "4": "B-bank_routing_number",
+  "5": "B-biometric_identifier",
+  "6": "B-blood_type",
+  "7": "B-certificate_license_number",
+  "8": "B-city",
+  "9": "B-company_name",
+  "10": "B-coordinate",
+  "11": "B-country",
+  "12": "B-county",
+  "13": "B-credit_debit_card",
+  "14": "B-customer_id",
+  "15": "B-cvv",
+  "16": "B-date",
+  "17": "B-date_of_birth",
+  "18": "B-date_time",
+  "19": "B-device_identifier",
+  "20": "B-education_level",
+  "21": "B-email",
+  "22": "B-employee_id",
+  "23": "B-employment_status",
+  "24": "B-fax_number",
+  "25": "B-first_name",
+  "26": "B-gender",
+  "27": "B-health_plan_beneficiary_number",
+  "28": "B-http_cookie",
+  "29": "B-ipv4",
+  "30": "B-ipv6",
+  "31": "B-language",
+  "32": "B-last_name",
+  "33": "B-license_plate",
+  "34": "B-mac_address",
+  "35": "B-medical_record_number",
+  "36": "B-occupation",
+  "37": "B-password",
+  "38": "B-phone_number",
+  "39": "B-pin",
+  "40": "B-political_view",
+  "41": "B-postcode",
+  "42": "B-race_ethnicity",
+  "43": "B-religious_belief",
+  "44": "B-sexuality",
+  "45": "B-ssn",
+  "46": "B-state",
+  "47": "B-street_address",
+  "48": "B-swift_bic",
+  "49": "B-tax_id",
+  "50": "B-time",
+  "51": "B-unique_id",
+  "52": "B-url",
+  "53": "B-user_name",
+  "54": "B-vehicle_identifier",
+  "55": "I-account_number",
+  "56": "I-api_key",
+  "57": "I-biometric_identifier",
+  "58": "I-blood_type",
+  "59": "I-certificate_license_number",
+  "60": "I-city",
+  "61": "I-company_name",
+  "62": "I-coordinate",
+  "63": "I-country",
+  "64": "I-county",
+  "65": "I-credit_debit_card",
+  "66": "I-customer_id",
+  "67": "I-date",
+  "68": "I-date_of_birth",
+  "69": "I-date_time",
+  "70": "I-device_identifier",
+  "71": "I-education_level",
+  "72": "I-email",
+  "73": "I-employee_id",
+  "74": "I-employment_status",
+  "75": "I-fax_number",
+  "76": "I-first_name",
+  "77": "I-gender",
+  "78": "I-health_plan_beneficiary_number",
+  "79": "I-http_cookie",
+  "80": "I-ipv4",
+  "81": "I-ipv6",
+  "82": "I-language",
+  "83": "I-last_name",
+  "84": "I-license_plate",
+  "85": "I-mac_address",
+  "86": "I-medical_record_number",
+  "87": "I-occupation",
+  "88": "I-password",
+  "89": "I-phone_number",
+  "90": "I-pin",
+  "91": "I-political_view",
+  "92": "I-postcode",
+  "93": "I-race_ethnicity",
+  "94": "I-religious_belief",
+  "95": "I-sexuality",
+  "96": "I-ssn",
+  "97": "I-state",
+  "98": "I-street_address",
+  "99": "I-swift_bic",
+  "100": "I-tax_id",
+  "101": "I-time",
+  "102": "I-unique_id",
+  "103": "I-url",
+  "104": "I-user_name",
+  "105": "I-vehicle_identifier"
+}

weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40960dd37e81743d08742d8a91c996f903df43047470d622551d422320efdd8b
+size 496567329