| { | |
| "module_name": "kmeans_quantize", | |
| "class_name": "KMeansCompressor", | |
| "category": "compression", | |
| "family": "compression", | |
| "params": [ | |
| "linear", | |
| "n_clusters", | |
| "model", | |
| "bits" | |
| ], | |
| "ops": [ | |
| "linear" | |
| ], | |
| "docstring": "KMeansCompressor: Vector quantization via scalar k-means clustering of weight parameters.\n\nArgs:\n model: PyTorch model.\n n_clusters: Number of centroid codebook entries (default 256 -> 8 bits, 16 -> 4 bits).\n bits: Optional bitwidth constraint (e.g. 4 or 8) which sets n_clusters = 2**bits.", | |
| "rank": 16, | |
| "d_state": 1, | |
| "num_experts": 1, | |
| "top_k": 1, | |
| "sparsity_pct": 0.0, | |
| "bits": 4, | |
| "hash_val": 33960237832199527375226833645264205257 | |
| } |
Xet Storage Details
- Size:
- 711 Bytes
- Xet hash:
- 2968e73b6f88ad85e228617e6684ec7c5f0f9ce13b2c91077cd1690edc9abf33
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.