whenxuan: upload the model config
Browse files- config.json +39 -0
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "FlowVortex/LiteSpecFormer",
|
| 3 |
+
"litespecformer_pipeline_class": "LiteSpecFormerPipeline",
|
| 4 |
+
"attn_implementation": "sdpa",
|
| 5 |
+
"attn_output_gate": "headwise",
|
| 6 |
+
"d_ff": 1536,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 512,
|
| 9 |
+
"dropout_rate": 0.1,
|
| 10 |
+
"feed_forward_proj": "gelu",
|
| 11 |
+
"initializer_factor": 0.05,
|
| 12 |
+
"is_encoder_decoder": false,
|
| 13 |
+
"kernel_size": 3,
|
| 14 |
+
"layer_norm_epsilon": 1e-06,
|
| 15 |
+
"n_lags": 48,
|
| 16 |
+
"num_heads": 8,
|
| 17 |
+
"num_layers": 6,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"reduction": 64,
|
| 20 |
+
"rope_theta": 10000.0,
|
| 21 |
+
"use_acf_loss": true,
|
| 22 |
+
"use_channel_attention": false,
|
| 23 |
+
"use_dw_cnn": false,
|
| 24 |
+
"vocab_size": 2,
|
| 25 |
+
"history_token_number": 3,
|
| 26 |
+
"forecasting_config": {
|
| 27 |
+
"context_length": 768,
|
| 28 |
+
"input_patch_size": 16,
|
| 29 |
+
"input_patch_stride": 16,
|
| 30 |
+
"max_output_patches": 1,
|
| 31 |
+
"medium_index": 5,
|
| 32 |
+
"min_past": 32,
|
| 33 |
+
"output_patch_size": 16,
|
| 34 |
+
"quantiles": [0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.95],
|
| 35 |
+
"time_encoding_scale": 768,
|
| 36 |
+
"use_arcsinh": true,
|
| 37 |
+
"use_reg_token": false
|
| 38 |
+
}
|
| 39 |
+
}
|