Upload folder using huggingface_hub

README.md

@@ -0,0 +1,90 @@
+---
+license: apache-2.0
+base_model: facebook/hubert-large-ll60k
+tags:
+- audio-classification
+- generated_from_trainer
+datasets:
+- superb
+metrics:
+- accuracy
+model-index:
+- name: superb_ks_42
+  results:
+  - task:
+      name: Audio Classification
+      type: audio-classification
+    dataset:
+      name: superb
+      type: superb
+      config: ks
+      split: validation
+      args: ks
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6209179170344219
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# superb_ks_42
+
+This model is a fine-tuned version of [facebook/hubert-large-ll60k](https://huggingface.co/facebook/hubert-large-ll60k) on the superb dataset.
+It achieves the following results on the evaluation set:
+- Loss: -11.1651
+- Accuracy: 0.6209
+- Test Accuracy: 0.6209
+- Df Accuracy: 0.1395
+- Unlearn Overall Accuracy: 0.7407
+- Unlearn Time: 720.5727
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Overall Accuracy | Unlearn Overall Accuracy | Time |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------------:|:------------------------:|:----:|
+| No log        | 1.0   | 64   | -4.3569         | 0.2168   | 0.7192           | 0.7192                   | -1   |
+| No log        | 2.0   | 128  | -11.1651        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 3.0   | 192  | -16.2146        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 4.0   | 256  | -21.2411        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 5.0   | 320  | -26.0470        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 6.0   | 384  | -30.3817        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 7.0   | 448  | -34.0174        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 8.0   | 512  | -36.7682        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 9.0   | 576  | -38.4767        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+| No log        | 10.0  | 640  | -39.0640        | 0.1395   | 0.7407           | 0.7407                   | -1   |
+
+
+### Framework versions
+
+- Transformers 4.39.3
+- Pytorch 2.2.2+cu118
+- Datasets 2.18.0
+- Tokenizers 0.15.2

all_results.json

@@ -0,0 +1,28 @@
+{
+    "df_accuracy": 0.1395007342143906,
+    "dr_accuracy": 0.6577847546431266,
+    "dr_loss": -10.081528663635254,
+    "dr_runtime": 175.3214,
+    "dr_samples_per_second": 279.778,
+    "dr_steps_per_second": 34.976,
+    "dr_unlearn_overall_accuracy": 0.759142010214368,
+    "epoch": 10.0,
+    "eval_accuracy": 0.6209179170344219,
+    "eval_loss": -11.165149688720703,
+    "eval_runtime": 27.8116,
+    "eval_samples_per_second": 244.43,
+    "eval_steps_per_second": 30.563,
+    "eval_unlearn_overall_accuracy": 0.7407085914100157,
+    "test_accuracy": 0.6577847546431266,
+    "test_loss": -27.74724578857422,
+    "test_runtime": 14.733,
+    "test_samples_per_second": 209.122,
+    "test_steps_per_second": 26.2,
+    "test_unlearn_overall_accuracy": 0.47195687080257426,
+    "train_loss": -53.131683349609375,
+    "train_runtime": 719.9884,
+    "train_samples_per_second": 28.375,
+    "train_steps_per_second": 0.889,
+    "unlearn_overall_accuracy": 0.759142010214368,
+    "unlearn_time": 720.5726902484894
+}

config.json

@@ -0,0 +1,112 @@
+{
+  "_name_or_path": "facebook/hubert-large-ll60k",
+  "activation_dropout": 0.0,
+  "apply_spec_augment": true,
+  "architectures": [
+    "HubertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "sum",
+  "ctc_zero_infinity": false,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.1,
+  "feat_proj_layer_norm": true,
+  "final_dropout": 0.0,
+  "finetuning_task": "audio-classification",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "yes",
+    "1": "no",
+    "10": "_silence_",
+    "11": "_unknown_",
+    "2": "up",
+    "3": "down",
+    "4": "left",
+    "5": "right",
+    "6": "on",
+    "7": "off",
+    "8": "stop",
+    "9": "go"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "_silence_": "10",
+    "_unknown_": "11",
+    "down": "3",
+    "go": "9",
+    "left": "4",
+    "no": "1",
+    "off": "7",
+    "on": "6",
+    "right": "5",
+    "stop": "8",
+    "up": "2",
+    "yes": "0"
+  },
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.075,
+  "mask_time_selection": "static",
+  "model_type": "hubert",
+  "num_attention_heads": 16,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32
+}

dr_results.json

@@ -0,0 +1,13 @@
+{
+    "df_accuracy": 0.1395007342143906,
+    "dr_accuracy": 0.6577847546431266,
+    "dr_loss": -10.081528663635254,
+    "dr_runtime": 175.3214,
+    "dr_samples_per_second": 279.778,
+    "dr_steps_per_second": 34.976,
+    "dr_unlearn_overall_accuracy": 0.759142010214368,
+    "epoch": 10.0,
+    "test_accuracy": 0.6577847546431266,
+    "unlearn_overall_accuracy": 0.759142010214368,
+    "unlearn_time": 720.5726902484894
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "df_accuracy": 0.1395007342143906,
+    "epoch": 10.0,
+    "eval_accuracy": 0.6209179170344219,
+    "eval_loss": -11.165149688720703,
+    "eval_runtime": 27.8116,
+    "eval_samples_per_second": 244.43,
+    "eval_steps_per_second": 30.563,
+    "eval_unlearn_overall_accuracy": 0.7407085914100157,
+    "test_accuracy": 0.6209179170344219,
+    "unlearn_overall_accuracy": 0.7407085914100157,
+    "unlearn_time": 720.5726902484894
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:824e08a95e8ab20446f7abf32a84f8eeaa53589d2b0558214a789c7d62fb1b19
+size 1262868744

pred_logit_df.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af4ea67ae486ec2cce63308bbbcb1e5c4b3d6e6ae4e01138e248053bfec8c750
+size 98192

pred_logit_dr.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0012464e98a339c876c8e13c9e0b0caa4c69570259d4afe79806d85aeb5fe6ea
+size 2354576

pred_logit_eval.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90f1c22ddf1023e13e4c7a9b842a68ae6686c984fb3dc25c5cf6b3147611ecce
+size 326432

pred_logit_test.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba8ff25539bf4823baba0305e0f2f9eb9a9c93d56251ac1d28943f4eb4987d7d
+size 148016

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

test_results.json

@@ -0,0 +1,12 @@
+{
+    "df_accuracy": 0.1395007342143906,
+    "epoch": 10.0,
+    "test_accuracy": 0.0834144758195391,
+    "test_loss": -27.74724578857422,
+    "test_runtime": 14.733,
+    "test_samples_per_second": 209.122,
+    "test_steps_per_second": 26.2,
+    "test_unlearn_overall_accuracy": 0.47195687080257426,
+    "unlearn_overall_accuracy": 0.47195687080257426,
+    "unlearn_time": 720.5726902484894
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 10.0,
+    "train_loss": -53.131683349609375,
+    "train_runtime": 719.9884,
+    "train_samples_per_second": 28.375,
+    "train_steps_per_second": 0.889
+}

trainer_state.json

@@ -0,0 +1,170 @@
+{
+  "best_metric": 0.7407085914100157,
+  "best_model_checkpoint": "audio/unlearn/checkpoint/hubert-large/neggrad/4.0/superb_ks_42/checkpoint-128",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 640,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "df_accuracy": 0.21683798335780716,
+      "epoch": 1.0,
+      "eval_accuracy": 0.655192703736393,
+      "eval_loss": -4.356904029846191,
+      "eval_runtime": 27.4955,
+      "eval_samples_per_second": 247.241,
+      "eval_steps_per_second": 30.914,
+      "eval_unlearn_overall_accuracy": 0.719177360189293,
+      "step": 64,
+      "test_accuracy": 0.655192703736393,
+      "unlearn_overall_accuracy": 0.719177360189293,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 2.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -11.165149688720703,
+      "eval_runtime": 29.0102,
+      "eval_samples_per_second": 234.332,
+      "eval_steps_per_second": 29.3,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 128,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 3.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -16.214649200439453,
+      "eval_runtime": 27.4869,
+      "eval_samples_per_second": 247.317,
+      "eval_steps_per_second": 30.924,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 192,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 4.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -21.24113655090332,
+      "eval_runtime": 27.3813,
+      "eval_samples_per_second": 248.272,
+      "eval_steps_per_second": 31.043,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 256,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 5.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -26.047012329101562,
+      "eval_runtime": 27.4946,
+      "eval_samples_per_second": 247.248,
+      "eval_steps_per_second": 30.915,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 320,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 6.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -30.381650924682617,
+      "eval_runtime": 27.2732,
+      "eval_samples_per_second": 249.255,
+      "eval_steps_per_second": 31.166,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 384,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 7.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -34.01741027832031,
+      "eval_runtime": 27.4944,
+      "eval_samples_per_second": 247.25,
+      "eval_steps_per_second": 30.915,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 448,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 8.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -36.768184661865234,
+      "eval_runtime": 27.4047,
+      "eval_samples_per_second": 248.06,
+      "eval_steps_per_second": 31.017,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 512,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 9.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -38.476722717285156,
+      "eval_runtime": 27.347,
+      "eval_samples_per_second": 248.583,
+      "eval_steps_per_second": 31.082,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 576,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "df_accuracy": 0.1395007342143906,
+      "epoch": 10.0,
+      "eval_accuracy": 0.6209179170344219,
+      "eval_loss": -39.06399917602539,
+      "eval_runtime": 27.6129,
+      "eval_samples_per_second": 246.19,
+      "eval_steps_per_second": 30.783,
+      "eval_unlearn_overall_accuracy": 0.7407085914100157,
+      "step": 640,
+      "test_accuracy": 0.6209179170344219,
+      "unlearn_overall_accuracy": 0.7407085914100157,
+      "unlearn_time": -1
+    },
+    {
+      "epoch": 10.0,
+      "step": 640,
+      "total_flos": 6.1918434122112e+17,
+      "train_loss": -53.131683349609375,
+      "train_runtime": 719.9884,
+      "train_samples_per_second": 28.375,
+      "train_steps_per_second": 0.889
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 640,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 6.1918434122112e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21fff055055af3e9d809190243372428906105df399a55690bc567b08899dea2
+size 5048