iitolstykh commited on
Commit
f08510a
·
verified ·
1 Parent(s): ff6d907

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +51 -0
README.md CHANGED
@@ -57,6 +57,57 @@ The model was evaluated on the test split of the `LLMTrace Detection dataset`. T
57
  | mAP @ IoU=0.5 | 0.8976 |
58
  | mAP @ IoU=0.5:0.95 | 0.7921 |
59
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
 
61
  ## Citation
62
 
 
57
  | mAP @ IoU=0.5 | 0.8976 |
58
  | mAP @ IoU=0.5:0.95 | 0.7921 |
59
 
60
+ ## Quick start
61
+
62
+ Requirements:
63
+ - python3.11
64
+ - [gigacheck](https://github.com/ai-forever/gigacheck)
65
+
66
+ ```bash
67
+ pip install git+https://github.com/ai-forever/gigacheck
68
+ ```
69
+
70
+ ### Inference with transformers (with trust_remote_code=True)
71
+
72
+ ```python
73
+ from transformers import AutoModel
74
+ import torch
75
+
76
+ model_name = "iitolstykh/GigaCheck-Detector-Multi"
77
+ gigacheck_model = AutoModel.from_pretrained(
78
+ model_name, trust_remote_code=True, device_map="cuda:0", torch_dtype=torch.float32
79
+ )
80
+
81
+ text = "The critic's review of the recent publication was scathing. The book failed miserably in portraying the harmful subjective discourses associated with the hegemony of the political system."
82
+
83
+ output = gigacheck_model([text], conf_interval_thresh=0.5)
84
+
85
+ # [(start_char, end_char, score)]
86
+ print(output.ai_intervals)
87
+ ```
88
+
89
+ ### Inference with gigacheck
90
+
91
+ ```python
92
+ from transformers import AutoConfig
93
+ from gigacheck.inference.src.mistral_detector import MistralDetector
94
+ import torch
95
+
96
+ model_name = "iitolstykh/GigaCheck-Detector-Multi"
97
+
98
+ config = AutoConfig.from_pretrained(model_name)
99
+ model = MistralDetector(
100
+ max_seq_len=config.max_length,
101
+ with_detr=config.with_detr,
102
+ id2label=config.id2label,
103
+ device="cpu" if not torch.cuda.is_available() else "cuda:0",
104
+ conf_interval_thresh=0.5,
105
+ ).from_pretrained(model_name)
106
+
107
+ text = "The critic's review of the recent publication was scathing. The book failed miserably in portraying the harmful subjective discourses associated with the hegemony of the political system."
108
+ output = model.predict(text)
109
+ print(output)
110
+ ```
111
 
112
  ## Citation
113