Upload folder using huggingface_hub

README.md

@@ -0,0 +1,70 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: dandelin/vilt-b32-finetuned-nlvr2
+tags:
+- image-text-classification
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: '42'
+  results: []
+---
+
+<!-- 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. -->
+
+# 42
+
+This model is a fine-tuned version of [dandelin/vilt-b32-finetuned-nlvr2](https://huggingface.co/dandelin/vilt-b32-finetuned-nlvr2) on the nlvr2 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5263
+- Accuracy: 0.7355
+- Dt Accuracy: 0.7355
+- Df Accuracy: 0.8759
+- Unlearn Overall Accuracy: 0.6218
+- Unlearn Time: None
+
+## 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: 1e-05
+- train_batch_size: 32
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 5
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Overall Accuracy | Unlearn Overall Accuracy | Time |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------------:|:------------------------:|:----:|
+| No log        | 1.0   | 324  | 0.5474          | 0.9126   | 0.5881           | 0.5881                   | None |
+| 0.4885        | 2.0   | 648  | 0.5345          | 0.9118   | 0.5892           | 0.5892                   | None |
+| 0.4885        | 3.0   | 972  | 0.5297          | 0.8911   | 0.6089           | 0.6089                   | None |
+| 0.3043        | 4.0   | 1296 | 0.5268          | 0.8842   | 0.6147           | 0.6147                   | None |
+| 0.2354        | 5.0   | 1620 | 0.5263          | 0.8759   | 0.6218           | 0.6218                   | None |
+
+
+### Framework versions
+
+- Transformers 4.48.0
+- Pytorch 2.3.0+cu121
+- Datasets 2.18.0
+- Tokenizers 0.21.0

all_results.json

@@ -0,0 +1,22 @@
+{
+    "df_accuracy": 0.8758700696055685,
+    "df_loss": 0.45554879307746887,
+    "dr_accuracy": 0.9296666666666666,
+    "dr_loss": 0.31367406249046326,
+    "dt_accuracy": 0.7475240419118703,
+    "epoch": 5.0,
+    "eval_unlearn_overall_accuracy": 0.621775694795363,
+    "knowledge_gap": 0.526869682907966,
+    "ood_accuracy": 0.748493543758967,
+    "ood_loss": 0.5186445116996765,
+    "test_accuracy": 0.7475240419118703,
+    "test_loss": 0.5181884169578552,
+    "total_flos": 1004374260677568.0,
+    "train_loss": 0.3335036242449725,
+    "train_runtime": 2328.2549,
+    "train_samples_per_second": 22.214,
+    "train_steps_per_second": 0.696,
+    "unlearn_overall_accuracy": 0.6242314492621494,
+    "unlearn_time": 2328.6262109279633,
+    "zrf": 0.9749588523016187
+}

config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "dandelin/vilt-b32-finetuned-nlvr2",
+  "architectures": [
+    "ViltForImagesAndTextClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "False",
+    "1": "True"
+  },
+  "image_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "False": 0,
+    "True": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_image_length": -1,
+  "max_position_embeddings": 40,
+  "modality_type_vocab_size": 3,
+  "model_type": "vilt",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "num_images": 2,
+  "patch_size": 32,
+  "qkv_bias": true,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.0",
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

df_results.json

@@ -0,0 +1,4 @@
+{
+    "df_accuracy": 0.8758700696055685,
+    "df_loss": 0.45554879307746887
+}

dr_results.json

@@ -0,0 +1,4 @@
+{
+    "dr_accuracy": 0.9296666666666666,
+    "dr_loss": 0.31367406249046326
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fa1f2d5fcf973c220356c800929dceffa95c2327f6fcb1c18040a417d808f37
+size 455875832

ood_results.json

@@ -0,0 +1,4 @@
+{
+    "ood_accuracy": 0.748493543758967,
+    "ood_loss": 0.5186445116996765
+}

pred_logit_df.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df4e9b27afba8846adcb28e73683ab121322ddb4a0665fcdd076af38c21de281
+size 41504

pred_logit_dr.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b30684be20a5ecc109e8d002ecb882dc650d66002b25ddd361b998237f73875
+size 48128

pred_logit_eval.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e50698328427f806cfd69879ee7199749e697d1d3b948a63c3dab8d9f65cccc3
+size 55984

pred_logit_ood.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e354ba167d09fd432afddc3b7d5637fb66940a88d61f94c63a84e9a3b99ad145
+size 55888

pred_logit_test.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73684eac19abc20c5fb30c624076ac046c3ee71d2eca3381c165cc848e536fbd
+size 55864

test_results.json

@@ -0,0 +1,4 @@
+{
+    "test_accuracy": 0.7475240419118703,
+    "test_loss": 0.5181884169578552
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "total_flos": 1004374260677568.0,
+    "train_loss": 0.3335036242449725,
+    "train_runtime": 2328.2549,
+    "train_samples_per_second": 22.214,
+    "train_steps_per_second": 0.696
+}

trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": 0.621775694795363,
+  "best_model_checkpoint": "../../checkpoint/unlearn/nlvr2/vilt/bad_teaching/6.0/42/checkpoint-1620",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 1620,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "df_accuracy": 0.9126063418406806,
+      "dt_accuracy": 0.7420509882555142,
+      "epoch": 1.0,
+      "eval_accuracy": 0.7420509882555142,
+      "eval_loss": 0.5473781228065491,
+      "eval_runtime": 113.5475,
+      "eval_samples_per_second": 61.49,
+      "eval_steps_per_second": 0.969,
+      "eval_unlearn_overall_accuracy": 0.5881464664707937,
+      "step": 324,
+      "unlearn_overall_accuracy": 0.5881464664707937,
+      "unlearn_time": null
+    },
+    {
+      "epoch": 1.5432098765432098,
+      "grad_norm": 2.551623582839966,
+      "learning_rate": 6.913580246913581e-06,
+      "loss": 0.4885,
+      "step": 500
+    },
+    {
+      "df_accuracy": 0.9118329466357309,
+      "dt_accuracy": 0.7437696934975652,
+      "epoch": 2.0,
+      "eval_accuracy": 0.7437696934975652,
+      "eval_loss": 0.5345158576965332,
+      "eval_runtime": 104.9313,
+      "eval_samples_per_second": 66.539,
+      "eval_steps_per_second": 1.048,
+      "eval_unlearn_overall_accuracy": 0.5892136585732513,
+      "step": 648,
+      "unlearn_overall_accuracy": 0.5892136585732513,
+      "unlearn_time": null
+    },
+    {
+      "df_accuracy": 0.8911446249033256,
+      "dt_accuracy": 0.7424806645660269,
+      "epoch": 3.0,
+      "eval_accuracy": 0.7424806645660269,
+      "eval_loss": 0.5296567678451538,
+      "eval_runtime": 119.9673,
+      "eval_samples_per_second": 58.199,
+      "eval_steps_per_second": 0.917,
+      "eval_unlearn_overall_accuracy": 0.6088830335218481,
+      "step": 972,
+      "unlearn_overall_accuracy": 0.6088830335218481,
+      "unlearn_time": null
+    },
+    {
+      "epoch": 3.0864197530864197,
+      "grad_norm": 2.1789159774780273,
+      "learning_rate": 3.827160493827161e-06,
+      "loss": 0.3043,
+      "step": 1000
+    },
+    {
+      "df_accuracy": 0.884184068058778,
+      "dt_accuracy": 0.7386135777714122,
+      "epoch": 4.0,
+      "eval_accuracy": 0.7386135777714122,
+      "eval_loss": 0.5267872214317322,
+      "eval_runtime": 95.2807,
+      "eval_samples_per_second": 73.278,
+      "eval_steps_per_second": 1.154,
+      "eval_unlearn_overall_accuracy": 0.6146807727893482,
+      "step": 1296,
+      "unlearn_overall_accuracy": 0.6146807727893482,
+      "unlearn_time": null
+    },
+    {
+      "epoch": 4.62962962962963,
+      "grad_norm": 3.276827096939087,
+      "learning_rate": 7.407407407407407e-07,
+      "loss": 0.2354,
+      "step": 1500
+    },
+    {
+      "df_accuracy": 0.8758700696055685,
+      "dt_accuracy": 0.7354626181609853,
+      "epoch": 5.0,
+      "eval_accuracy": 0.7354626181609853,
+      "eval_loss": 0.5263105630874634,
+      "eval_runtime": 87.2124,
+      "eval_samples_per_second": 80.057,
+      "eval_steps_per_second": 1.261,
+      "eval_unlearn_overall_accuracy": 0.621775694795363,
+      "step": 1620,
+      "unlearn_overall_accuracy": 0.621775694795363,
+      "unlearn_time": null
+    },
+    {
+      "epoch": 5.0,
+      "step": 1620,
+      "total_flos": 1004374260677568.0,
+      "train_loss": 0.3335036242449725,
+      "train_runtime": 2328.2549,
+      "train_samples_per_second": 22.214,
+      "train_steps_per_second": 0.696
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1620,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1004374260677568.0,
+  "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:c11f0bc61c040d43fe8355c9644f5ac3cbd2b5c2dd741b85880c5166906b1f7e
+size 5496

unlearn_final_results.json

@@ -0,0 +1,11 @@
+{
+    "df_accuracy": 0.8758700696055685,
+    "dr_accuracy": 0.9296666666666666,
+    "dt_accuracy": 0.7475240419118703,
+    "eval_unlearn_overall_accuracy": 0.621775694795363,
+    "knowledge_gap": 0.526869682907966,
+    "ood_accuracy": 0.748493543758967,
+    "unlearn_overall_accuracy": 0.6242314492621494,
+    "unlearn_time": 2328.6262109279633,
+    "zrf": 0.9749588523016187
+}