hf-tiny-v2/tiny-random-PatchTSMixerForRegression

1{2  "architectures": [3    "PatchTSMixerForRegression"4  ],5  "channel_consistent_masking": true,6  "context_length": 32,7  "d_model": 8,8  "distribution_output": "student_t",9  "dropout": 0.5,10  "dtype": "float32",11  "expansion_factor": 2,12  "forecast_mask_ratios": [13    1,14    115  ],16  "gated_attn": true,17  "head_aggregation": null,18  "head_dropout": 0.2,19  "init_std": 0.02,20  "loss": "mse",21  "mask_mode": "mask_before_encoder",22  "mask_patches": [23    2,24    325  ],26  "mask_type": "forecast",27  "mask_value": 0,28  "masked_loss": false,29  "mode": "common_channel",30  "model_type": "patchtsmixer",31  "norm_eps": 1e-05,32  "norm_mlp": "LayerNorm",33  "num_forecast_mask_patches": [34    235  ],36  "num_input_channels": 3,37  "num_layers": 2,38  "num_parallel_samples": 100,39  "num_patches": 4,40  "num_targets": 3,41  "out_channels": null,42  "output_range": null,43  "patch_last": true,44  "patch_length": 8,45  "patch_stride": 8,46  "positional_encoding_type": "sincos",47  "prediction_channel_indices": null,48  "prediction_length": 16,49  "random_mask_ratio": 0.5,50  "scaling": "std",51  "self_attn": false,52  "self_attn_heads": 1,53  "swin_hier": 0,54  "transformers_version": "5.16.0.dev0",55  "unmasked_channel_indices": null,56  "use_positional_encoding": false57}58