Upload folder using huggingface_hub
Browse files- config.json +27 -0
- confusion_matrix.png +0 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- training_metrics.json +81 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"eos_token_id": null,
|
| 12 |
+
"hidden_dim": 3072,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "distilbert",
|
| 16 |
+
"n_heads": 12,
|
| 17 |
+
"n_layers": 6,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"qa_dropout": 0.1,
|
| 20 |
+
"seq_classif_dropout": 0.2,
|
| 21 |
+
"sinusoidal_pos_embds": false,
|
| 22 |
+
"tie_weights_": true,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"transformers_version": "5.2.0",
|
| 25 |
+
"use_cache": false,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
confusion_matrix.png
ADDED
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdfd635c0df927e6f5c1038da96f8cf4535803724bf1b949c399464ccb5b615f
|
| 3 |
+
size 267832560
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": true,
|
| 5 |
+
"is_local": false,
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 512,
|
| 8 |
+
"pad_token": "[PAD]",
|
| 9 |
+
"sep_token": "[SEP]",
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "BertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
training_metrics.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dataset_stats": {
|
| 3 |
+
"total_samples": 8438,
|
| 4 |
+
"train_size": 7594,
|
| 5 |
+
"val_size": 844,
|
| 6 |
+
"class_distribution_raw": {
|
| 7 |
+
"0": 0.7250533301730268,
|
| 8 |
+
"1": 0.2749466698269732
|
| 9 |
+
}
|
| 10 |
+
},
|
| 11 |
+
"training_metrics": {
|
| 12 |
+
"best_f1": 0.989247311827957,
|
| 13 |
+
"best_model_accuracy": 0.9940758293838863
|
| 14 |
+
},
|
| 15 |
+
"calibration": {
|
| 16 |
+
"optimal_threshold": 0.0027917588595300913,
|
| 17 |
+
"calibrated_precision": 0.9508196721311475,
|
| 18 |
+
"calibrated_recall": 1.0,
|
| 19 |
+
"class_weights_used": {
|
| 20 |
+
"0": 0.689604445897352,
|
| 21 |
+
"1": 1.8185344827586207
|
| 22 |
+
}
|
| 23 |
+
},
|
| 24 |
+
"robustness": [
|
| 25 |
+
{
|
| 26 |
+
"type": "URL Obfuscation",
|
| 27 |
+
"text": "Win $1000 fast! Click hxxp://scammy-link[.]com",
|
| 28 |
+
"probability": 0.9997965693473816,
|
| 29 |
+
"predicted": 1,
|
| 30 |
+
"expected": 1,
|
| 31 |
+
"passed": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"type": "URL Obfuscation",
|
| 35 |
+
"text": "Your account is locked. Verify at bit.ly/secure-login-392",
|
| 36 |
+
"probability": 0.9997977614402771,
|
| 37 |
+
"predicted": 1,
|
| 38 |
+
"expected": 1,
|
| 39 |
+
"passed": true
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"type": "Numeric Replacement",
|
| 43 |
+
"text": "W1NNER! Y0u have b33n select3d for a pr1ze.",
|
| 44 |
+
"probability": 0.9936283230781555,
|
| 45 |
+
"predicted": 1,
|
| 46 |
+
"expected": 1,
|
| 47 |
+
"passed": true
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"type": "Mixed Case",
|
| 51 |
+
"text": "cOnGrAtUlAtIoNs, yOu WoN a FrEe iPhOnE!",
|
| 52 |
+
"probability": 0.8983756899833679,
|
| 53 |
+
"predicted": 1,
|
| 54 |
+
"expected": 1,
|
| 55 |
+
"passed": true
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"type": "Short SMS",
|
| 59 |
+
"text": "Call me now",
|
| 60 |
+
"probability": 0.00036344866384752095,
|
| 61 |
+
"predicted": 0,
|
| 62 |
+
"expected": 1,
|
| 63 |
+
"passed": false
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"type": "Short SMS",
|
| 67 |
+
"text": "Hey",
|
| 68 |
+
"probability": 0.0003704441187437624,
|
| 69 |
+
"predicted": 0,
|
| 70 |
+
"expected": 0,
|
| 71 |
+
"passed": true
|
| 72 |
+
}
|
| 73 |
+
],
|
| 74 |
+
"deployment": {
|
| 75 |
+
"model_size_mb": 255.41,
|
| 76 |
+
"cpu_latency_estimate": "~10-30ms / sequence",
|
| 77 |
+
"gpu_latency_estimate": "~2-5ms / sequence",
|
| 78 |
+
"recommendation": "CPU (FastAPI wrapped, ONNX quantization recommended)"
|
| 79 |
+
},
|
| 80 |
+
"model_path": "./cipher_distilbert_detection"
|
| 81 |
+
}
|