Upload folder using huggingface_hub

finetuned_ttm_model/config.json

@@ -0,0 +1,66 @@
+{
+  "adaptive_patching_levels": 0,
+  "architectures": [
+    "TinyTimeMixerForPrediction"
+  ],
+  "categorical_vocab_size_list": null,
+  "context_length": 52,
+  "d_model": 20,
+  "d_model_scale": 5,
+  "decoder_adaptive_patching_levels": 0,
+  "decoder_d_model": 16,
+  "decoder_d_model_scale": 4,
+  "decoder_mode": "mix_channel",
+  "decoder_num_layers": 4,
+  "decoder_raw_residual": false,
+  "distribution_output": "student_t",
+  "dropout": 0.3,
+  "enable_forecast_channel_mixing": false,
+  "exogenous_channel_indices": null,
+  "expansion_factor": 3,
+  "fcm_context_length": 1,
+  "fcm_gated_attn": true,
+  "fcm_mix_layers": 3,
+  "fcm_prepend_past": true,
+  "fcm_prepend_past_offset": null,
+  "fcm_use_mixer": true,
+  "frequency_token_vocab_size": 10,
+  "gated_attn": true,
+  "head_dropout": 0.3,
+  "huber_delta": 1,
+  "init_embed": "pytorch",
+  "init_linear": "pytorch",
+  "init_processing": true,
+  "init_std": 0.02,
+  "loss": "mse",
+  "mask_value": 0,
+  "masked_context_length": null,
+  "mode": "common_channel",
+  "model_type": "tinytimemixer",
+  "norm_eps": 1e-05,
+  "norm_mlp": "LayerNorm",
+  "num_input_channels": 8,
+  "num_layers": 12,
+  "num_parallel_samples": 100,
+  "num_patches": 14,
+  "patch_last": true,
+  "patch_length": 4,
+  "patch_stride": 4,
+  "positional_encoding_type": "sincos",
+  "post_init": false,
+  "prediction_channel_indices": [
+    0
+  ],
+  "prediction_filter_length": 6,
+  "prediction_length": 16,
+  "quantile": 0.5,
+  "resolution_prefix_tuning": true,
+  "scaling": "std",
+  "self_attn": false,
+  "self_attn_heads": 1,
+  "stride_ratio": 1,
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "use_decoder": true,
+  "use_positional_encoding": false
+}

finetuned_ttm_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a697a662117fd92d5a872b07456e7d507f2d7cf51db5030baa015bd4d18ce501
+size 332048

logs/events.out.tfevents.1745270714.dbfd7c3c91c7.1623.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a21c50543671d323491ff23170c9b9a258b56102d0b72350d1f4a9af71008b62
+size 12124

output/checkpoint-2786/config.json

@@ -0,0 +1,66 @@
+{
+  "adaptive_patching_levels": 0,
+  "architectures": [
+    "TinyTimeMixerForPrediction"
+  ],
+  "categorical_vocab_size_list": null,
+  "context_length": 52,
+  "d_model": 20,
+  "d_model_scale": 5,
+  "decoder_adaptive_patching_levels": 0,
+  "decoder_d_model": 16,
+  "decoder_d_model_scale": 4,
+  "decoder_mode": "mix_channel",
+  "decoder_num_layers": 4,
+  "decoder_raw_residual": false,
+  "distribution_output": "student_t",
+  "dropout": 0.3,
+  "enable_forecast_channel_mixing": false,
+  "exogenous_channel_indices": null,
+  "expansion_factor": 3,
+  "fcm_context_length": 1,
+  "fcm_gated_attn": true,
+  "fcm_mix_layers": 3,
+  "fcm_prepend_past": true,
+  "fcm_prepend_past_offset": null,
+  "fcm_use_mixer": true,
+  "frequency_token_vocab_size": 10,
+  "gated_attn": true,
+  "head_dropout": 0.3,
+  "huber_delta": 1,
+  "init_embed": "pytorch",
+  "init_linear": "pytorch",
+  "init_processing": true,
+  "init_std": 0.02,
+  "loss": "mse",
+  "mask_value": 0,
+  "masked_context_length": null,
+  "mode": "common_channel",
+  "model_type": "tinytimemixer",
+  "norm_eps": 1e-05,
+  "norm_mlp": "LayerNorm",
+  "num_input_channels": 8,
+  "num_layers": 12,
+  "num_parallel_samples": 100,
+  "num_patches": 14,
+  "patch_last": true,
+  "patch_length": 4,
+  "patch_stride": 4,
+  "positional_encoding_type": "sincos",
+  "post_init": false,
+  "prediction_channel_indices": [
+    0
+  ],
+  "prediction_filter_length": 6,
+  "prediction_length": 16,
+  "quantile": 0.5,
+  "resolution_prefix_tuning": true,
+  "scaling": "std",
+  "self_attn": false,
+  "self_attn_heads": 1,
+  "stride_ratio": 1,
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "use_decoder": true,
+  "use_positional_encoding": false
+}

output/checkpoint-2786/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a697a662117fd92d5a872b07456e7d507f2d7cf51db5030baa015bd4d18ce501
+size 332048

output/checkpoint-2786/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5de32065dbd7d9e934710939301d37c86ed9519d38bbe7684d87e3c798f467be
+size 811194

output/checkpoint-2786/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dd30c2f71a549d40d88501427dd6c934beddaec3a87220f6f567c9139ce997e
+size 13990

output/checkpoint-2786/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28758e9bf3a0e44359800d4da07f861242d93e71b6cd40bf30e1f8d873dba782
+size 1320

output/checkpoint-2786/trainer_state.json

@@ -0,0 +1,148 @@
+{
+  "best_global_step": 2786,
+  "best_metric": 72.45570373535156,
+  "best_model_checkpoint": "model/output/checkpoint-2786",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 2786,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1758.0869140625,
+      "learning_rate": 0.00033533537238222464,
+      "loss": 165.6237,
+      "step": 398
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 76.47968292236328,
+      "eval_runtime": 20.0392,
+      "eval_samples_per_second": 202.802,
+      "eval_steps_per_second": 3.194,
+      "step": 398
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 661.80322265625,
+      "learning_rate": 0.0009673161575995766,
+      "loss": 101.9303,
+      "step": 796
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 74.72091674804688,
+      "eval_runtime": 16.517,
+      "eval_samples_per_second": 246.05,
+      "eval_steps_per_second": 3.875,
+      "step": 796
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 233.8292999267578,
+      "learning_rate": 0.0016381468554296547,
+      "loss": 96.1758,
+      "step": 1194
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 76.53939819335938,
+      "eval_runtime": 16.7409,
+      "eval_samples_per_second": 242.759,
+      "eval_steps_per_second": 3.823,
+      "step": 1194
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2881.05810546875,
+      "learning_rate": 0.001989264776725461,
+      "loss": 92.2926,
+      "step": 1592
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 74.74008178710938,
+      "eval_runtime": 18.2222,
+      "eval_samples_per_second": 223.025,
+      "eval_steps_per_second": 3.512,
+      "step": 1592
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1436.7108154296875,
+      "learning_rate": 0.001967294993859577,
+      "loss": 91.7396,
+      "step": 1990
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 74.24183654785156,
+      "eval_runtime": 16.4508,
+      "eval_samples_per_second": 247.04,
+      "eval_steps_per_second": 3.89,
+      "step": 1990
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1180.649169921875,
+      "learning_rate": 0.0018380887525394208,
+      "loss": 89.8515,
+      "step": 2388
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 79.67572021484375,
+      "eval_runtime": 16.8586,
+      "eval_samples_per_second": 241.063,
+      "eval_steps_per_second": 3.796,
+      "step": 2388
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1331.4036865234375,
+      "learning_rate": 0.0016234913078995259,
+      "loss": 87.4604,
+      "step": 2786
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 72.45570373535156,
+      "eval_runtime": 16.5884,
+      "eval_samples_per_second": 244.991,
+      "eval_steps_per_second": 3.858,
+      "step": 2786
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 5572,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 14,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 32459312820480.0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

output/checkpoint-2786/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6232fa152053185effc28232db54f86c5b5e5f051938087ee56ad17a1212bf3
+size 5240