End of training

README.md

@@ -0,0 +1,76 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: facebook/deit-tiny-distilled-patch16-224
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: deit-tiny-distilled-patch16-224emotion_model_binary_deit
+  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. -->
+
+# deit-tiny-distilled-patch16-224emotion_model_binary_deit
+
+This model is a fine-tuned version of [facebook/deit-tiny-distilled-patch16-224](https://huggingface.co/facebook/deit-tiny-distilled-patch16-224) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7254
+- Accuracy: 0.9056
+- Weighted f1: 0.9056
+- Micro f1: 0.9056
+- Macro f1: 0.9056
+- Weighted recall: 0.9056
+- Micro recall: 0.9056
+- Macro recall: 0.9056
+
+## 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: 64
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:--------:|:---------------:|:------------:|:------------:|
+| 0.4537        | 1.0   | 401  | 0.3859          | 0.8234   | 0.8219      | 0.8234   | 0.8219   | 0.8234          | 0.8234       | 0.8234       |
+| 0.3044        | 2.0   | 802  | 0.3653          | 0.8422   | 0.8411      | 0.8422   | 0.8411   | 0.8422          | 0.8422       | 0.8422       |
+| 0.1886        | 3.0   | 1203 | 0.2977          | 0.8859   | 0.8859      | 0.8859   | 0.8859   | 0.8859          | 0.8859       | 0.8859       |
+| 0.093         | 4.0   | 1604 | 0.3351          | 0.8972   | 0.8972      | 0.8972   | 0.8972   | 0.8972          | 0.8972       | 0.8972       |
+| 0.048         | 5.0   | 2005 | 0.4311          | 0.9025   | 0.9025      | 0.9025   | 0.9025   | 0.9025          | 0.9025       | 0.9025       |
+| 0.0245        | 6.0   | 2406 | 0.5580          | 0.9034   | 0.9034      | 0.9034   | 0.9034   | 0.9034          | 0.9034       | 0.9034       |
+| 0.0101        | 7.0   | 2807 | 0.6712          | 0.9044   | 0.9044      | 0.9044   | 0.9044   | 0.9044          | 0.9044       | 0.9044       |
+| 0.0029        | 8.0   | 3208 | 0.7049          | 0.9041   | 0.9041      | 0.9041   | 0.9041   | 0.9041          | 0.9041       | 0.9041       |
+| 0.0011        | 9.0   | 3609 | 0.7212          | 0.9047   | 0.9047      | 0.9047   | 0.9047   | 0.9047          | 0.9047       | 0.9047       |
+| 0.0006        | 10.0  | 4010 | 0.7254          | 0.9056   | 0.9056      | 0.9056   | 0.9056   | 0.9056          | 0.9056       | 0.9056       |
+
+
+### Framework versions
+
+- Transformers 4.57.1
+- Pytorch 2.8.0+cu126
+- Datasets 4.0.0
+- Tokenizers 0.22.1

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.2780826449901978e+18,
+    "train_loss": 0.11275497220103581,
+    "train_runtime": 1741.1173,
+    "train_samples_per_second": 147.101,
+    "train_steps_per_second": 2.303
+}

config.json

@@ -0,0 +1,33 @@
+{
+  "architectures": [
+    "DeiTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "dtype": "float32",
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 192,
+  "id2label": {
+    "0": 0,
+    "1": 1
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
+  "label2id": {
+    "0": "0",
+    "1": "1"
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "deit",
+  "num_attention_heads": 3,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "pooler_act": "tanh",
+  "pooler_output_size": 192,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "transformers_version": "4.57.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c981b2ebaabbc5b0725f4d221583b06b7fda43d0260c2616c2d22b98bf3608b
+size 22123752

preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "DeiTFeatureExtractor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

runs/Nov02_16-42-47_9b6325b29d09/events.out.tfevents.1762101771.9b6325b29d09.932.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffa9c65fd5149f94fe92f832deb27a760226055bf4f2ad6de6f021b9bdf9afe6
+size 14347

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.2780826449901978e+18,
+    "train_loss": 0.11275497220103581,
+    "train_runtime": 1741.1173,
+    "train_samples_per_second": 147.101,
+    "train_steps_per_second": 2.303
+}

trainer_state.json

@@ -0,0 +1,270 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 4010,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0024937655860349127,
+      "grad_norm": 5.275404930114746,
+      "learning_rate": 5e-05,
+      "loss": 0.6955,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 8.534090995788574,
+      "learning_rate": 4.5012468827930175e-05,
+      "loss": 0.4537,
+      "step": 401
+    },
+    {
+      "epoch": 1.0,
+      "eval_Macro F1": 0.8219241158379686,
+      "eval_Macro Recall": 0.8234375,
+      "eval_Micro F1": 0.8234375,
+      "eval_Micro Recall": 0.8234375,
+      "eval_Weighted F1": 0.8219241158379686,
+      "eval_Weighted Recall": 0.8234375,
+      "eval_accuracy": 0.8234375,
+      "eval_loss": 0.3859266936779022,
+      "eval_runtime": 17.4087,
+      "eval_samples_per_second": 183.816,
+      "eval_steps_per_second": 22.977,
+      "step": 401
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 18.81427001953125,
+      "learning_rate": 4.0012468827930175e-05,
+      "loss": 0.3044,
+      "step": 802
+    },
+    {
+      "epoch": 2.0,
+      "eval_Macro F1": 0.8410977627331971,
+      "eval_Macro Recall": 0.8421875000000001,
+      "eval_Micro F1": 0.8421875,
+      "eval_Micro Recall": 0.8421875,
+      "eval_Weighted F1": 0.8410977627331971,
+      "eval_Weighted Recall": 0.8421875,
+      "eval_accuracy": 0.8421875,
+      "eval_loss": 0.36532944440841675,
+      "eval_runtime": 15.4731,
+      "eval_samples_per_second": 206.811,
+      "eval_steps_per_second": 25.851,
+      "step": 802
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 9.516457557678223,
+      "learning_rate": 3.5012468827930175e-05,
+      "loss": 0.1886,
+      "step": 1203
+    },
+    {
+      "epoch": 3.0,
+      "eval_Macro F1": 0.8859012981755637,
+      "eval_Macro Recall": 0.8859375,
+      "eval_Micro F1": 0.8859375,
+      "eval_Micro Recall": 0.8859375,
+      "eval_Weighted F1": 0.8859012981755637,
+      "eval_Weighted Recall": 0.8859375,
+      "eval_accuracy": 0.8859375,
+      "eval_loss": 0.2977246046066284,
+      "eval_runtime": 14.811,
+      "eval_samples_per_second": 216.056,
+      "eval_steps_per_second": 27.007,
+      "step": 1203
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9479053616523743,
+      "learning_rate": 3.0012468827930173e-05,
+      "loss": 0.093,
+      "step": 1604
+    },
+    {
+      "epoch": 4.0,
+      "eval_Macro F1": 0.8971671644050704,
+      "eval_Macro Recall": 0.8971875,
+      "eval_Micro F1": 0.8971875,
+      "eval_Micro Recall": 0.8971875,
+      "eval_Weighted F1": 0.8971671644050704,
+      "eval_Weighted Recall": 0.8971875,
+      "eval_accuracy": 0.8971875,
+      "eval_loss": 0.3351382315158844,
+      "eval_runtime": 15.5296,
+      "eval_samples_per_second": 206.058,
+      "eval_steps_per_second": 25.757,
+      "step": 1604
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.17879733443260193,
+      "learning_rate": 2.5012468827930173e-05,
+      "loss": 0.048,
+      "step": 2005
+    },
+    {
+      "epoch": 5.0,
+      "eval_Macro F1": 0.9024981337533413,
+      "eval_Macro Recall": 0.9025,
+      "eval_Micro F1": 0.9025,
+      "eval_Micro Recall": 0.9025,
+      "eval_Weighted F1": 0.9024981337533413,
+      "eval_Weighted Recall": 0.9025,
+      "eval_accuracy": 0.9025,
+      "eval_loss": 0.43113386631011963,
+      "eval_runtime": 14.6907,
+      "eval_samples_per_second": 217.825,
+      "eval_steps_per_second": 27.228,
+      "step": 2005
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.9963405132293701,
+      "learning_rate": 2.0012468827930177e-05,
+      "loss": 0.0245,
+      "step": 2406
+    },
+    {
+      "epoch": 6.0,
+      "eval_Macro F1": 0.9034129665162021,
+      "eval_Macro Recall": 0.9034375,
+      "eval_Micro F1": 0.9034375,
+      "eval_Micro Recall": 0.9034375,
+      "eval_Weighted F1": 0.903412966516202,
+      "eval_Weighted Recall": 0.9034375,
+      "eval_accuracy": 0.9034375,
+      "eval_loss": 0.5580075979232788,
+      "eval_runtime": 14.5513,
+      "eval_samples_per_second": 219.911,
+      "eval_steps_per_second": 27.489,
+      "step": 2406
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.016222944483160973,
+      "learning_rate": 1.5012468827930176e-05,
+      "loss": 0.0101,
+      "step": 2807
+    },
+    {
+      "epoch": 7.0,
+      "eval_Macro F1": 0.9043552359061697,
+      "eval_Macro Recall": 0.9043749999999999,
+      "eval_Micro F1": 0.904375,
+      "eval_Micro Recall": 0.904375,
+      "eval_Weighted F1": 0.9043552359061696,
+      "eval_Weighted Recall": 0.904375,
+      "eval_accuracy": 0.904375,
+      "eval_loss": 0.6712324023246765,
+      "eval_runtime": 14.1596,
+      "eval_samples_per_second": 225.996,
+      "eval_steps_per_second": 28.249,
+      "step": 2807
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.004565633367747068,
+      "learning_rate": 1.0012468827930176e-05,
+      "loss": 0.0029,
+      "step": 3208
+    },
+    {
+      "epoch": 8.0,
+      "eval_Macro F1": 0.9040622657770161,
+      "eval_Macro Recall": 0.9040625,
+      "eval_Micro F1": 0.9040625,
+      "eval_Micro Recall": 0.9040625,
+      "eval_Weighted F1": 0.904062265777016,
+      "eval_Weighted Recall": 0.9040625,
+      "eval_accuracy": 0.9040625,
+      "eval_loss": 0.7048962116241455,
+      "eval_runtime": 14.9861,
+      "eval_samples_per_second": 213.532,
+      "eval_steps_per_second": 26.691,
+      "step": 3208
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.004040522966533899,
+      "learning_rate": 5.0124688279301756e-06,
+      "loss": 0.0011,
+      "step": 3609
+    },
+    {
+      "epoch": 9.0,
+      "eval_Macro F1": 0.9046863737354709,
+      "eval_Macro Recall": 0.9046875,
+      "eval_Micro F1": 0.9046875,
+      "eval_Micro Recall": 0.9046875,
+      "eval_Weighted F1": 0.904686373735471,
+      "eval_Weighted Recall": 0.9046875,
+      "eval_accuracy": 0.9046875,
+      "eval_loss": 0.7212379574775696,
+      "eval_runtime": 14.1678,
+      "eval_samples_per_second": 225.865,
+      "eval_steps_per_second": 28.233,
+      "step": 3609
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.014547404833137989,
+      "learning_rate": 1.2468827930174563e-08,
+      "loss": 0.0006,
+      "step": 4010
+    },
+    {
+      "epoch": 10.0,
+      "eval_Macro F1": 0.9056240783601401,
+      "eval_Macro Recall": 0.9056249999999999,
+      "eval_Micro F1": 0.905625,
+      "eval_Micro Recall": 0.905625,
+      "eval_Weighted F1": 0.9056240783601401,
+      "eval_Weighted Recall": 0.905625,
+      "eval_accuracy": 0.905625,
+      "eval_loss": 0.7254420518875122,
+      "eval_runtime": 14.2344,
+      "eval_samples_per_second": 224.808,
+      "eval_steps_per_second": 28.101,
+      "step": 4010
+    },
+    {
+      "epoch": 10.0,
+      "step": 4010,
+      "total_flos": 1.2780826449901978e+18,
+      "train_loss": 0.11275497220103581,
+      "train_runtime": 1741.1173,
+      "train_samples_per_second": 147.101,
+      "train_steps_per_second": 2.303
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "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": 1.2780826449901978e+18,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c82c0c0ac4fef20982f39637a503c1c6cf892918ba1e005f7a8d6c91e614a0bc
+size 5905

validation_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.2780826449901978e+18,
+    "train_loss": 0.11275497220103581,
+    "train_runtime": 1741.1173,
+    "train_samples_per_second": 147.101,
+    "train_steps_per_second": 2.303
+}