ise-uiuc / titanfuzz Goto Github PK

I am a student currently studying this project, and I am using coverage.py to collect line coverage data. However, the results I obtained are quite different from those in the paper, which has left me feeling confused.

How to obtain possible defect information from trace.txt? Is it through manual screening? like

When I try to run bash scripts/demo_run_tf.sh, I encounter "Cannot set a non-string value as the PAD token". How can I resolve this?

Complete Output:

Current directory:  /home/src/run
Results will be dumped to:  /home/src/run/Results
[0 / 2] tf.nn.conv2d
output.returncode:  0
stdout>  Current directory:  /home/src/run
Results will be dumped to:  /home/src/run/Results
api:  tf.nn.conv2d
Initializing a SpanLM based model: facebook/incoder-1B ...
Max length: 2048
Cannot set a non-string value as the PAD token

(titanfuzz) [tly@localhost TitanFuzz]$ bash scripts/demo_run_torch.sh false
Warning: running in a non-docker environment!
Current directory: /home/tly/Fuzzing/TitanFuzz/TitanFuzz
Results will be dumped to: /home/tly/Fuzzing/TitanFuzz/TitanFuzz/Results
[0 / 2] torch.nn.RReLU
output.returncode: 0
stdout> Current directory: /home/tly/Fuzzing/TitanFuzz/TitanFuzz
Results will be dumped to: /home/tly/Fuzzing/TitanFuzz/TitanFuzz/Results

api: torch.nn.RReLU
Initializing a SpanLM based model: facebook/incoder-1B ...
pretrained: facebook/incoder-1B
Model loaded from: /home/tly/Fuzzing/TitanFuzz/TitanFuzz/LLM/incoder-1B
Model configuration: XGLMConfig {
"_name_or_path": "/home/tly/Fuzzing/TitanFuzz/TitanFuzz/LLM/incoder-1B",
"activation_dropout": 0.0,
"activation_function": "gelu",
"architectures": [
"XGLMForCausalLM"
],
"attention_dropout": 0.1,
"attention_heads": 32,
"bos_token_id": 0,
"d_model": 2048,
"decoder_start_token_id": 2,
"dropout": 0.1,
"eos_token_id": 2,
"ffn_dim": 8192,
"init_std": 0.02,
"layerdrop": 0.0,
"max_position_embeddings": 2048,
"model_type": "xglm",
"num_layers": 24,
"pad_token_id": 1,
"scale_embedding": true,
"torch_dtype": "float32",
"transformers_version": "4.39.3",
"use_cache": true,
"vocab_size": 50518
}

Max length: 2048
Batch size: 30
Model loading time: 45.33568048477173
'Constant' object has no attribute 'kind'

[1 / 2] torch.nn.Conv2d
output.returncode: 0
stdout> Current directory: /home/tly/Fuzzing/TitanFuzz/TitanFuzz
Results will be dumped to: /home/tly/Fuzzing/TitanFuzz/TitanFuzz/Results

api: torch.nn.Conv2d
Initializing a SpanLM based model: facebook/incoder-1B ...
pretrained: facebook/incoder-1B
Model loaded from: /home/tly/Fuzzing/TitanFuzz/TitanFuzz/LLM/incoder-1B
Model configuration: XGLMConfig {
"_name_or_path": "/home/tly/Fuzzing/TitanFuzz/TitanFuzz/LLM/incoder-1B",
"activation_dropout": 0.0,
"activation_function": "gelu",
"architectures": [
"XGLMForCausalLM"
],
"attention_dropout": 0.1,
"attention_heads": 32,
"bos_token_id": 0,
"d_model": 2048,
"decoder_start_token_id": 2,
"dropout": 0.1,
"eos_token_id": 2,
"ffn_dim": 8192,
"init_std": 0.02,
"layerdrop": 0.0,
"max_position_embeddings": 2048,
"model_type": "xglm",
"num_layers": 24,
"pad_token_id": 1,
"scale_embedding": true,
"torch_dtype": "float32",
"transformers_version": "4.39.3",
"use_cache": true,
"vocab_size": 50518
}

Max length: 2048
Batch size: 30
Model loading time: 41.063090324401855
'Constant' object has no attribute 'kind'

这个错误是由于incoder-1B导致的吗