Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model: microsoft/unixcoder-base-nine
|
| 4 |
+
tags:
|
| 5 |
+
- code
|
| 6 |
+
- ai-generated-code-detection
|
| 7 |
+
- classifier
|
| 8 |
+
library_name: transformers
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# code-detection-confound checkpoints
|
| 12 |
+
|
| 13 |
+
Three fine-tuned **AI-generated-code detection** classifiers from the
|
| 14 |
+
[`AMindToThink/code-detection-confound`](https://github.com/AMindToThink/code-detection-confound)
|
| 15 |
+
research project. All three are cross-entropy-only (CE) fine-tunes of
|
| 16 |
+
[`microsoft/unixcoder-base-nine`](https://huggingface.co/microsoft/unixcoder-base-nine);
|
| 17 |
+
they differ only in training data.
|
| 18 |
+
|
| 19 |
+
| Subfolder | Training data |
|
| 20 |
+
|---|---|
|
| 21 |
+
| `unixcoder_dc_ce/` | DroidCollection-Python |
|
| 22 |
+
| `python_raw_ce/` | HMCorp / Python |
|
| 23 |
+
| `java_raw_ce/` | HMCorp / Java |
|
| 24 |
+
|
| 25 |
+
Each `model.bin` (~481 MB) is a **raw PyTorch `state_dict`** — no `config.json` or tokenizer
|
| 26 |
+
is bundled. Load it on top of the `microsoft/unixcoder-base-nine` architecture + tokenizer.
|
| 27 |
+
The exact training command (`scripts/18_train_cgs_amp.py … --model_name_or_path
|
| 28 |
+
microsoft/unixcoder-base-nine`), data provenance, and the classification head are documented
|
| 29 |
+
in the source repo's `REPRODUCE.md`.
|
| 30 |
+
|
| 31 |
+
Backed up here during a machine migration (2026-07-02); see the source repo for full
|
| 32 |
+
reproduction details.
|