update the transformer lib cannot detect

README.md

@@ -14,13 +14,27 @@ pipeline_tag: time-series-classification
 
 Time-RCD is a transformer-based model for zero-shot anomaly detection in time series data.
 
+## ⚠️ IMPORTANT: Custom Model Loading
+
+**This model uses a custom architecture not built into transformers.**  
+You **MUST** include `trust_remote_code=True` when loading:
+
+```python
+# ✅ CORRECT - Will work
+from transformers import AutoModel
+model = AutoModel.from_pretrained("your-repo/Time-RCD", trust_remote_code=True)
+
+# ❌ WRONG - Will throw KeyError: 'time_rcd'
+model = AutoModel.from_pretrained("your-repo/Time-RCD")  # Missing trust_remote_code=True
+```
+
 ## Quick Start
 
 ```python
 from transformers import AutoModel, AutoConfig
 import numpy as np
 
-# Load model
+# Load model (trust_remote_code=True is REQUIRED!)
 model = AutoModel.from_pretrained(
     "thu-sail-lab/Time_RCD",
     trust_remote_code=True
@@ -72,35 +86,44 @@ anomaly_scores = outputs.anomaly_scores.numpy()
 from transformers import AutoModel
 import numpy as np
 
+# IMPORTANT: trust_remote_code=True is required!
 model = AutoModel.from_pretrained("thu-sail-lab/Time_RCD", trust_remote_code=True)
 
 # Your time series (n_samples, n_features)
 data = np.random.randn(10000, 1)
 
-# Get anomaly scores
-outputs = model.zero_shot(data)
-scores = outputs.anomaly_scores.numpy()
+# Get anomaly scores using the zero_shot method
+scores, logits = model.zero_shot(data)
+
+# Flatten scores from list of batches
+import numpy as np
+all_scores = np.concatenate(scores, axis=0).flatten()
 
 # Detect anomalies (e.g., top 5%)
-threshold = np.percentile(scores, 95)
-anomalies = scores > threshold
+threshold = np.percentile(all_scores, 95)
+anomalies = all_scores > threshold
 ```
 
 ### With Custom Processing
 
 ```python
-from transformers import AutoModel, AutoProcessor
+from transformers import AutoModel
+from processing_time_rcd import TimeRCDProcessor
+import numpy as np
 
+# IMPORTANT: trust_remote_code=True is required!
 model = AutoModel.from_pretrained("thu-sail-lab/Time_RCD", trust_remote_code=True)
-processor = AutoProcessor.from_pretrained("thu-sail-lab/Time_RCD", trust_remote_code=True)
 
-# Configure processor
-processor.win_size = 5000
-processor.normalize = True
+# Create and configure processor
+processor = TimeRCDProcessor(win_size=5000, normalize=True)
 
-# Process and detect
+# Process data
+data = np.random.randn(10000, 1)
 processed = processor(data, return_tensors="pt")
+
+# Get predictions
 outputs = model(**processed)
+anomaly_scores = outputs.anomaly_scores
 ```
 
 ## Configuration
@@ -124,13 +147,50 @@ Evaluated on various time series anomaly detection benchmarks.
 - Performance varies by domain
 - High-dimensional data may need preprocessing
 
+## Troubleshooting
+
+### KeyError: 'time_rcd'
+
+If you see this error:
+```
+KeyError: 'time_rcd'
+The checkpoint you are trying to load has model type `time_rcd` but Transformers does not recognize this architecture.
+```
+
+**Solution:** Add `trust_remote_code=True` to your loading code:
+
+```python
+# This will fix the error
+model = AutoModel.from_pretrained("your-repo/Time-RCD", trust_remote_code=True)
+```
+
+This is required because Time-RCD is a custom architecture not built into the transformers library. The `trust_remote_code=True` flag tells transformers to load and execute the custom model code from the repository.
+
+### Other Common Issues
+
+**Issue:** `ModuleNotFoundError: No module named 'einops'`  
+**Solution:** Install einops: `pip install einops`
+
+**Issue:** Model runs slowly on CPU  
+**Solution:** Move model to GPU: `model = model.to('cuda')`
+
+**Issue:** Out of memory errors  
+**Solution:** Reduce window size or batch size in the processor:
+```python
+processor = TimeRCDProcessor(win_size=2000, batch_size=32)
+```
+
 ## Citation
 
 ```bibtex
-@article{time-rcd,
-  title={Time-RCD: Zero-Shot Time Series Anomaly Detection},
-  author={Your Name},
-  year={2025}
+@misc{lan2025foundationmodelszeroshottime,
+      title={Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy}, 
+      author={Tian Lan and Hao Duong Le and Jinbo Li and Wenjun He and Meng Wang and Chenghao Liu and Chen Zhang},
+      year={2025},
+      eprint={2509.21190},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/2509.21190}, 
 }
 ```
 

__init__.py

@@ -0,0 +1,23 @@
+"""
+Time-RCD: Zero-Shot Time Series Anomaly Detection
+
+This package provides the Time-RCD model for zero-shot anomaly detection in time series data.
+
+Usage:
+    >>> from transformers import AutoModel
+    >>> model = AutoModel.from_pretrained("your-repo/Time-RCD", trust_remote_code=True)
+"""
+
+from .configuration_time_rcd import TimeRCDConfig
+from .modeling_time_rcd import Time_RCD, TimeRCDOutput
+from .processing_time_rcd import TimeRCDProcessor
+
+__version__ = "1.0.0"
+
+__all__ = [
+    "TimeRCDConfig",
+    "Time_RCD",
+    "TimeRCDOutput",
+    "TimeRCDProcessor",
+]
+

configuration_time_rcd.py

@@ -102,3 +102,10 @@ class TimeRCDConfig(PretrainedConfig):
 
 # Backward compatibility alias
 AnomalyCLIPConfig = TimeRCDConfig
+
+# Register config with AutoConfig when using trust_remote_code
+try:
+    from transformers import AutoConfig
+    AutoConfig.register("time_rcd", TimeRCDConfig)
+except Exception:
+    pass  # Silently fail if already registered or in restricted environment

modeling_time_rcd.py

@@ -738,3 +738,10 @@ class LlamaMLP(nn.Module):
 # For backward compatibility, create aliases
 TimeRCDModel = Time_RCD  # Alias for consistency
 AnomalyCLIPModel = Time_RCD  # For existing code that uses this name
+
+# Register model with AutoModel when using trust_remote_code
+try:
+    from transformers import AutoModel
+    AutoModel.register(TimeRCDConfig, Time_RCD)
+except Exception:
+    pass  # Silently fail if already registered or in restricted environment