Upload manual_upload with huggingface_hub

manual_upload/manual_upload/manual_upload/manual_upload/.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

manual_upload/manual_upload/manual_upload/manual_upload/.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

manual_upload/manual_upload/manual_upload/manual_upload/special_tokens_map.json

@@ -0,0 +1,3 @@
+{
+  "pad_token": "[PAD]"
+}

manual_upload/manual_upload/manual_upload/manual_upload/tokenizer.json

@@ -0,0 +1,357 @@
+{
+  "version": "1.0",
+  "truncation": {
+    "direction": "Right",
+    "max_length": 2048,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
+  "padding": {
+    "strategy": "BatchLongest",
+    "direction": "Right",
+    "pad_to_multiple_of": null,
+    "pad_id": 1,
+    "pad_type_id": 0,
+    "pad_token": "[PAD]"
+  },
+  "added_tokens": [
+    {
+      "id": 0,
+      "content": "[UNK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 1,
+      "content": "[PAD]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 2,
+      "content": "[MASK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  ],
+  "normalizer": null,
+  "pre_tokenizer": {
+    "type": "WhitespaceSplit"
+  },
+  "post_processor": null,
+  "decoder": null,
+  "model": {
+    "type": "WordLevel",
+    "vocab": {
+      "[UNK]": 0,
+      "[PAD]": 1,
+      "[MASK]": 2,
+      "TIME_DELTA=1": 3,
+      "TIME_DELTA=2": 4,
+      "BAR_END": 5,
+      "BAR_START": 6,
+      "NOTE_OFF=42": 7,
+      "NOTE_ON=42": 8,
+      "NOTE_OFF=36": 9,
+      "NOTE_ON=36": 10,
+      "TIME_DELTA=4": 11,
+      "NOTE_OFF=38": 12,
+      "NOTE_ON=38": 13,
+      "NOTE_OFF=54": 14,
+      "NOTE_ON=54": 15,
+      "NOTE_OFF=62": 16,
+      "NOTE_ON=62": 17,
+      "NOTE_OFF=64": 18,
+      "NOTE_ON=64": 19,
+      "TIME_DELTA=3": 20,
+      "NOTE_OFF=57": 21,
+      "NOTE_ON=57": 22,
+      "NOTE_OFF=69": 23,
+      "NOTE_ON=69": 24,
+      "TRACK_END": 25,
+      "TRACK_START": 26,
+      "NOTE_OFF=40": 27,
+      "NOTE_ON=40": 28,
+      "NOTE_OFF=60": 29,
+      "NOTE_ON=60": 30,
+      "NOTE_OFF=35": 31,
+      "NOTE_ON=35": 32,
+      "NOTE_OFF=59": 33,
+      "NOTE_ON=59": 34,
+      "NOTE_OFF=55": 35,
+      "NOTE_ON=55": 36,
+      "NOTE_OFF=46": 37,
+      "NOTE_ON=46": 38,
+      "NOTE_OFF=67": 39,
+      "NOTE_ON=67": 40,
+      "NOTE_OFF=70": 41,
+      "NOTE_ON=70": 42,
+      "NOTE_OFF=50": 43,
+      "NOTE_ON=50": 44,
+      "NOTE_OFF=44": 45,
+      "NOTE_ON=44": 46,
+      "NOTE_OFF=52": 47,
+      "NOTE_ON=52": 48,
+      "NOTE_OFF=61": 49,
+      "NOTE_ON=61": 50,
+      "NOTE_OFF=65": 51,
+      "NOTE_ON=65": 52,
+      "NOTE_OFF=63": 53,
+      "NOTE_ON=63": 54,
+      "NOTE_OFF=66": 55,
+      "NOTE_ON=66": 56,
+      "NOTE_OFF=45": 57,
+      "NOTE_ON=45": 58,
+      "NOTE_OFF=51": 59,
+      "NOTE_ON=51": 60,
+      "NOTE_OFF=43": 61,
+      "NOTE_ON=43": 62,
+      "NOTE_OFF=48": 63,
+      "NOTE_ON=48": 64,
+      "NOTE_OFF=58": 65,
+      "NOTE_ON=58": 66,
+      "NOTE_OFF=39": 67,
+      "NOTE_ON=39": 68,
+      "NOTE_OFF=53": 69,
+      "NOTE_ON=53": 70,
+      "NOTE_OFF=56": 71,
+      "NOTE_ON=56": 72,
+      "NOTE_OFF=47": 73,
+      "NOTE_ON=47": 74,
+      "NOTE_OFF=68": 75,
+      "NOTE_ON=68": 76,
+      "NOTE_OFF=49": 77,
+      "NOTE_ON=49": 78,
+      "NOTE_OFF=72": 79,
+      "NOTE_ON=72": 80,
+      "NOTE_OFF=71": 81,
+      "NOTE_ON=71": 82,
+      "NOTE_OFF=41": 83,
+      "NOTE_ON=41": 84,
+      "NOTE_OFF=74": 85,
+      "NOTE_ON=74": 86,
+      "NOTE_OFF=33": 87,
+      "NOTE_ON=33": 88,
+      "TIME_DELTA=6": 89,
+      "NOTE_OFF=82": 90,
+      "NOTE_ON=82": 91,
+      "TIME_DELTA=16": 92,
+      "TIME_DELTA=8": 93,
+      "NOTE_OFF=37": 94,
+      "NOTE_ON=37": 95,
+      "NOTE_OFF=31": 96,
+      "NOTE_ON=31": 97,
+      "NOTE_OFF=76": 98,
+      "NOTE_ON=76": 99,
+      "DENSITY=3": 100,
+      "NOTE_OFF=73": 101,
+      "NOTE_ON=73": 102,
+      "DENSITY=0": 103,
+      "NOTE_OFF=28": 104,
+      "NOTE_ON=28": 105,
+      "DENSITY=1": 106,
+      "DENSITY=2": 107,
+      "NOTE_OFF=34": 108,
+      "NOTE_ON=34": 109,
+      "INST=3": 110,
+      "NOTE_OFF=75": 111,
+      "NOTE_ON=75": 112,
+      "NOTE_OFF=77": 113,
+      "NOTE_ON=77": 114,
+      "PIECE_START": 115,
+      "NOTE_OFF=79": 116,
+      "NOTE_ON=79": 117,
+      "INST=DRUMS": 118,
+      "NOTE_OFF=32": 119,
+      "NOTE_ON=32": 120,
+      "NOTE_OFF=29": 121,
+      "NOTE_ON=29": 122,
+      "INST=4": 123,
+      "NOTE_OFF=81": 124,
+      "NOTE_ON=81": 125,
+      "TIME_DELTA=5": 126,
+      "NOTE_OFF=78": 127,
+      "NOTE_ON=78": 128,
+      "NOTE_OFF=30": 129,
+      "NOTE_ON=30": 130,
+      "NOTE_OFF=27": 131,
+      "NOTE_ON=27": 132,
+      "INST=6": 133,
+      "NOTE_OFF=80": 134,
+      "NOTE_ON=80": 135,
+      "TIME_DELTA=7": 136,
+      "NOTE_OFF=26": 137,
+      "NOTE_ON=26": 138,
+      "INST=0": 139,
+      "NOTE_OFF=83": 140,
+      "NOTE_ON=83": 141,
+      "TIME_DELTA=12": 142,
+      "TIME_DELTA=10": 143,
+      "NOTE_OFF=84": 144,
+      "NOTE_ON=84": 145,
+      "NOTE_OFF=86": 146,
+      "NOTE_ON=86": 147,
+      "INST=10": 148,
+      "NOTE_OFF=85": 149,
+      "NOTE_ON=85": 150,
+      "TIME_DELTA=14": 151,
+      "TIME_DELTA=15": 152,
+      "NOTE_OFF=88": 153,
+      "NOTE_ON=88": 154,
+      "INST=8": 155,
+      "INST=11": 156,
+      "NOTE_OFF=87": 157,
+      "NOTE_ON=87": 158,
+      "TIME_DELTA=9": 159,
+      "NOTE_OFF=24": 160,
+      "NOTE_ON=24": 161,
+      "INST=7": 162,
+      "NOTE_OFF=25": 163,
+      "NOTE_ON=25": 164,
+      "NOTE_OFF=89": 165,
+      "NOTE_ON=89": 166,
+      "NOTE_OFF=91": 167,
+      "NOTE_ON=91": 168,
+      "TIME_DELTA=11": 169,
+      "TIME_DELTA=13": 170,
+      "INST=2": 171,
+      "NOTE_OFF=93": 172,
+      "NOTE_ON=93": 173,
+      "NOTE_OFF=22": 174,
+      "NOTE_ON=22": 175,
+      "NOTE_OFF=23": 176,
+      "NOTE_ON=23": 177,
+      "NOTE_OFF=90": 178,
+      "NOTE_ON=90": 179,
+      "INST=9": 180,
+      "INST=5": 181,
+      "INST=1": 182,
+      "NOTE_OFF=94": 183,
+      "NOTE_ON=94": 184,
+      "INST=12": 185,
+      "INST=14": 186,
+      "NOTE_OFF=92": 187,
+      "NOTE_ON=92": 188,
+      "NOTE_OFF=96": 189,
+      "NOTE_ON=96": 190,
+      "NOTE_OFF=95": 191,
+      "NOTE_ON=95": 192,
+      "NOTE_OFF=98": 193,
+      "NOTE_ON=98": 194,
+      "INST=15": 195,
+      "NOTE_OFF=21": 196,
+      "NOTE_ON=21": 197,
+      "INST=13": 198,
+      "NOTE_OFF=19": 199,
+      "NOTE_ON=19": 200,
+      "NOTE_OFF=99": 201,
+      "NOTE_ON=99": 202,
+      "NOTE_OFF=97": 203,
+      "NOTE_ON=97": 204,
+      "NOTE_OFF=0": 205,
+      "NOTE_ON=0": 206,
+      "NOTE_OFF=100": 207,
+      "NOTE_ON=100": 208,
+      "NOTE_OFF=16": 209,
+      "NOTE_ON=16": 210,
+      "NOTE_OFF=13": 211,
+      "NOTE_ON=13": 212,
+      "NOTE_OFF=20": 213,
+      "NOTE_ON=20": 214,
+      "NOTE_OFF=105": 215,
+      "NOTE_ON=105": 216,
+      "NOTE_OFF=103": 217,
+      "NOTE_ON=103": 218,
+      "NOTE_OFF=101": 219,
+      "NOTE_ON=101": 220,
+      "NOTE_OFF=102": 221,
+      "NOTE_ON=102": 222,
+      "NOTE_OFF=17": 223,
+      "NOTE_ON=17": 224,
+      "NOTE_OFF=107": 225,
+      "NOTE_ON=107": 226,
+      "NOTE_OFF=108": 227,
+      "NOTE_ON=108": 228,
+      "NOTE_OFF=18": 229,
+      "NOTE_ON=18": 230,
+      "NOTE_OFF=126": 231,
+      "NOTE_ON=126": 232,
+      "NOTE_OFF=104": 233,
+      "NOTE_ON=104": 234,
+      "NOTE_OFF=8": 235,
+      "NOTE_ON=8": 236,
+      "NOTE_OFF=117": 237,
+      "NOTE_ON=117": 238,
+      "NOTE_OFF=106": 239,
+      "NOTE_ON=106": 240,
+      "NOTE_OFF=110": 241,
+      "NOTE_ON=110": 242,
+      "NOTE_OFF=112": 243,
+      "NOTE_ON=112": 244,
+      "NOTE_OFF=12": 245,
+      "NOTE_ON=12": 246,
+      "NOTE_OFF=9": 247,
+      "NOTE_ON=9": 248,
+      "NOTE_OFF=14": 249,
+      "NOTE_ON=14": 250,
+      "NOTE_OFF=113": 251,
+      "NOTE_ON=113": 252,
+      "NOTE_OFF=15": 253,
+      "NOTE_ON=15": 254,
+      "NOTE_OFF=125": 255,
+      "NOTE_ON=125": 256,
+      "NOTE_OFF=109": 257,
+      "NOTE_ON=109": 258,
+      "NOTE_OFF=115": 259,
+      "NOTE_ON=115": 260,
+      "NOTE_OFF=120": 261,
+      "NOTE_ON=120": 262,
+      "NOTE_OFF=119": 263,
+      "NOTE_ON=119": 264,
+      "NOTE_OFF=122": 265,
+      "NOTE_ON=122": 266,
+      "NOTE_OFF=124": 267,
+      "NOTE_OFF=127": 268,
+      "NOTE_ON=124": 269,
+      "NOTE_ON=127": 270,
+      "NOTE_OFF=11": 271,
+      "NOTE_ON=11": 272,
+      "NOTE_OFF=4": 273,
+      "NOTE_ON=4": 274,
+      "NOTE_OFF=10": 275,
+      "NOTE_ON=10": 276,
+      "NOTE_OFF=111": 277,
+      "NOTE_ON=111": 278,
+      "NOTE_OFF=5": 279,
+      "NOTE_OFF=6": 280,
+      "NOTE_ON=5": 281,
+      "NOTE_ON=6": 282,
+      "NOTE_OFF=1": 283,
+      "NOTE_ON=1": 284,
+      "NOTE_OFF=114": 285,
+      "NOTE_ON=114": 286,
+      "NOTE_OFF=2": 287,
+      "NOTE_ON=2": 288,
+      "NOTE_OFF=7": 289,
+      "NOTE_ON=7": 290,
+      "NOTE_OFF=3": 291,
+      "NOTE_ON=3": 292,
+      "NOTE_OFF=116": 293,
+      "NOTE_OFF=121": 294,
+      "NOTE_ON=116": 295,
+      "NOTE_ON=121": 296,
+      "NOTE_OFF=118": 297,
+      "NOTE_ON=118": 298
+    },
+    "unk_token": "[UNK]"
+  }
+}

manual_upload/manual_upload/manual_upload/manual_upload/tokenizer_config.json

@@ -0,0 +1,4 @@
+{
+  "model_max_length": 1000000000000000019884624838656,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

manual_upload/manual_upload/manual_upload/manual_upload/trainer_state.json

@@ -0,0 +1,15 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": null,
+  "global_step": 0,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [],
+  "max_steps": 0,
+  "num_train_epochs": 0,
+  "total_flos": 0,
+  "trial_name": null,
+  "trial_params": null
+}

manual_upload/manual_upload/manual_upload/manual_upload/training_args.json

@@ -0,0 +1,109 @@
+{
+  "output_dir": "models/improved_4bars",
+  "overwrite_output_dir": true,
+  "do_train": false,
+  "do_eval": true,
+  "do_predict": false,
+  "evaluation_strategy": "steps",
+  "prediction_loss_only": false,
+  "per_device_train_batch_size": 10,
+  "per_device_eval_batch_size": 8,
+  "per_gpu_train_batch_size": null,
+  "per_gpu_eval_batch_size": null,
+  "gradient_accumulation_steps": 1,
+  "eval_accumulation_steps": null,
+  "eval_delay": 0,
+  "learning_rate": 0.0005,
+  "weight_decay": 0.1,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "num_train_epochs": 10,
+  "max_steps": -1,
+  "lr_scheduler_type": "cosine",
+  "warmup_ratio": 0.0,
+  "warmup_steps": 5000,
+  "log_level": "passive",
+  "log_level_replica": "passive",
+  "log_on_each_node": true,
+  "logging_dir": "models/improved_4bars/logs",
+  "logging_strategy": "steps",
+  "logging_first_step": false,
+  "logging_steps": 2048,
+  "logging_nan_inf_filter": true,
+  "save_strategy": "steps",
+  "save_steps": 8192,
+  "save_total_limit": 5,
+  "save_on_each_node": false,
+  "no_cuda": false,
+  "use_mps_device": false,
+  "seed": 42,
+  "data_seed": null,
+  "jit_mode_eval": false,
+  "use_ipex": false,
+  "bf16": false,
+  "fp16": true,
+  "fp16_opt_level": "O1",
+  "half_precision_backend": "cuda_amp",
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "local_rank": -1,
+  "xpu_backend": null,
+  "tpu_num_cores": null,
+  "tpu_metrics_debug": false,
+  "debug": [],
+  "dataloader_drop_last": false,
+  "eval_steps": 2048,
+  "dataloader_num_workers": 0,
+  "past_index": -1,
+  "run_name": "models/improved_4bars",
+  "disable_tqdm": false,
+  "remove_unused_columns": true,
+  "label_names": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": null,
+  "greater_is_better": null,
+  "ignore_data_skip": false,
+  "sharded_ddp": [],
+  "fsdp": [],
+  "fsdp_min_num_params": 0,
+  "fsdp_transformer_layer_cls_to_wrap": null,
+  "deepspeed": null,
+  "label_smoothing_factor": 0.0,
+  "optim": "adamw_hf",
+  "optim_args": null,
+  "adafactor": false,
+  "group_by_length": false,
+  "length_column_name": "length",
+  "report_to": [
+    "wandb"
+  ],
+  "ddp_find_unused_parameters": null,
+  "ddp_bucket_cap_mb": null,
+  "dataloader_pin_memory": true,
+  "skip_memory_metrics": true,
+  "use_legacy_prediction_loop": false,
+  "push_to_hub": true,
+  "resume_from_checkpoint": null,
+  "hub_model_id": "JammyMachina/improved_4bars-mdl",
+  "hub_strategy": "every_save",
+  "hub_token": "<HUB_TOKEN>",
+  "hub_private_repo": false,
+  "gradient_checkpointing": false,
+  "include_inputs_for_metrics": false,
+  "fp16_backend": "auto",
+  "push_to_hub_model_id": null,
+  "push_to_hub_organization": null,
+  "push_to_hub_token": "<PUSH_TO_HUB_TOKEN>",
+  "mp_parameters": "",
+  "auto_find_batch_size": false,
+  "full_determinism": false,
+  "torchdynamo": null,
+  "ray_scope": "last",
+  "ddp_timeout": 1800,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null
+}

manual_upload/manual_upload/manual_upload/training_args.json

@@ -6,7 +6,7 @@
   "do_predict": false,
   "evaluation_strategy": "steps",
   "prediction_loss_only": false,
-  "per_device_train_batch_size": 10,
+  "per_device_train_batch_size": 8,
   "per_device_eval_batch_size": 8,
   "per_gpu_train_batch_size": null,
   "per_gpu_eval_batch_size": null,

manual_upload/manual_upload/training_args.json

@@ -6,7 +6,7 @@
   "do_predict": false,
   "evaluation_strategy": "steps",
   "prediction_loss_only": false,
-  "per_device_train_batch_size": 8,
+  "per_device_train_batch_size": 7,
   "per_device_eval_batch_size": 8,
   "per_gpu_train_batch_size": null,
   "per_gpu_eval_batch_size": null,

manual_upload/training_args.json

@@ -19,7 +19,7 @@
   "adam_beta2": 0.999,
   "adam_epsilon": 1e-08,
   "max_grad_norm": 1.0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 8,
   "max_steps": -1,
   "lr_scheduler_type": "cosine",
   "warmup_ratio": 0.0,