pushing roberta-large trained on Schone Power dataset

README.md

@@ -1,3 +1,68 @@
----
-license: apache-2.0
----
+---
+language:
+- en
+license: mit
+base_model: roberta-large
+tags:
+- generated_from_trainer
+datasets:
+- glue
+model-index:
+- name: final
+  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. -->
+
+# final
+
+This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the GLUE SCHONE_POW dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1556
+- Roc Auc: 0.9742
+
+## 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: 128
+- eval_batch_size: 256
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 256
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 5.0
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Roc Auc |
+|:-------------:|:------:|:----:|:---------------:|:-------:|
+| 0.2924        | 0.9985 | 332  | 0.2605          | 0.9248  |
+| 0.2553        | 2.0    | 665  | 0.2234          | 0.9468  |
+| 0.2317        | 2.9985 | 997  | 0.1899          | 0.9620  |
+| 0.2063        | 4.0    | 1330 | 0.1645          | 0.9715  |
+| 0.1897        | 4.9925 | 1660 | 0.1556          | 0.9742  |
+
+
+### Framework versions
+
+- Transformers 4.44.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 4.992481203007519,
+    "eval_loss": 0.155641108751297,
+    "eval_roc_auc": 0.9741751333869398,
+    "eval_runtime": 36.862,
+    "eval_samples": 16301,
+    "eval_samples_per_second": 442.217,
+    "eval_steps_per_second": 1.736,
+    "total_flos": 4.323837898114577e+16,
+    "train_loss": 0.24427399233163122,
+    "train_runtime": 2915.5625,
+    "train_samples": 85027,
+    "train_samples_per_second": 145.816,
+    "train_steps_per_second": 0.569
+}

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "roberta-large",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "schone_pow",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "No": 0,
+    "Yes": 1
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.992481203007519,
+    "eval_loss": 0.155641108751297,
+    "eval_roc_auc": 0.9741751333869398,
+    "eval_runtime": 36.862,
+    "eval_samples": 16301,
+    "eval_samples_per_second": 442.217,
+    "eval_steps_per_second": 1.736
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:208923dac2c8cc9ca874bf5e9f039e4963522c1d1d68120b9bf250ec84ee5dac
+size 1421495416

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.992481203007519,
+    "total_flos": 4.323837898114577e+16,
+    "train_loss": 0.24427399233163122,
+    "train_runtime": 2915.5625,
+    "train_samples": 85027,
+    "train_samples_per_second": 145.816,
+    "train_steps_per_second": 0.569
+}

trainer_state.json

@@ -0,0 +1,318 @@
+{
+  "best_metric": 0.155641108751297,
+  "best_model_checkpoint": "/cronus_data/avirinchipur/implicit_motives/schone_pow_roba_large/final/checkpoint-1660",
+  "epoch": 4.992481203007519,
+  "eval_steps": 500,
+  "global_step": 1660,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.15037593984962405,
+      "grad_norm": 7.807601451873779,
+      "learning_rate": 9.698795180722893e-06,
+      "loss": 0.4607,
+      "step": 50
+    },
+    {
+      "epoch": 0.3007518796992481,
+      "grad_norm": 4.938343524932861,
+      "learning_rate": 9.397590361445785e-06,
+      "loss": 0.383,
+      "step": 100
+    },
+    {
+      "epoch": 0.45112781954887216,
+      "grad_norm": 4.443939208984375,
+      "learning_rate": 9.096385542168675e-06,
+      "loss": 0.3231,
+      "step": 150
+    },
+    {
+      "epoch": 0.6015037593984962,
+      "grad_norm": 3.3444550037384033,
+      "learning_rate": 8.795180722891567e-06,
+      "loss": 0.3008,
+      "step": 200
+    },
+    {
+      "epoch": 0.7518796992481203,
+      "grad_norm": 6.133885383605957,
+      "learning_rate": 8.493975903614459e-06,
+      "loss": 0.2952,
+      "step": 250
+    },
+    {
+      "epoch": 0.9022556390977443,
+      "grad_norm": 4.0399627685546875,
+      "learning_rate": 8.19277108433735e-06,
+      "loss": 0.2924,
+      "step": 300
+    },
+    {
+      "epoch": 0.9984962406015038,
+      "eval_loss": 0.26054325699806213,
+      "eval_roc_auc": 0.9248130559023198,
+      "eval_runtime": 37.9704,
+      "eval_samples_per_second": 429.309,
+      "eval_steps_per_second": 1.686,
+      "step": 332
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 3.2617785930633545,
+      "learning_rate": 7.891566265060243e-06,
+      "loss": 0.2848,
+      "step": 350
+    },
+    {
+      "epoch": 1.2030075187969924,
+      "grad_norm": 2.823629140853882,
+      "learning_rate": 7.590361445783133e-06,
+      "loss": 0.2591,
+      "step": 400
+    },
+    {
+      "epoch": 1.3533834586466165,
+      "grad_norm": 8.734086990356445,
+      "learning_rate": 7.289156626506025e-06,
+      "loss": 0.2651,
+      "step": 450
+    },
+    {
+      "epoch": 1.5037593984962405,
+      "grad_norm": 5.4109601974487305,
+      "learning_rate": 6.987951807228917e-06,
+      "loss": 0.2658,
+      "step": 500
+    },
+    {
+      "epoch": 1.6541353383458648,
+      "grad_norm": 3.0360822677612305,
+      "learning_rate": 6.686746987951808e-06,
+      "loss": 0.2609,
+      "step": 550
+    },
+    {
+      "epoch": 1.8045112781954886,
+      "grad_norm": 4.336048603057861,
+      "learning_rate": 6.385542168674699e-06,
+      "loss": 0.2629,
+      "step": 600
+    },
+    {
+      "epoch": 1.954887218045113,
+      "grad_norm": 3.214564800262451,
+      "learning_rate": 6.084337349397591e-06,
+      "loss": 0.2553,
+      "step": 650
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.2233881801366806,
+      "eval_roc_auc": 0.9468388936522675,
+      "eval_runtime": 37.4563,
+      "eval_samples_per_second": 435.2,
+      "eval_steps_per_second": 1.709,
+      "step": 665
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 3.9632742404937744,
+      "learning_rate": 5.783132530120482e-06,
+      "loss": 0.2392,
+      "step": 700
+    },
+    {
+      "epoch": 2.255639097744361,
+      "grad_norm": 3.3193111419677734,
+      "learning_rate": 5.4819277108433745e-06,
+      "loss": 0.2336,
+      "step": 750
+    },
+    {
+      "epoch": 2.406015037593985,
+      "grad_norm": 3.099203109741211,
+      "learning_rate": 5.180722891566266e-06,
+      "loss": 0.2195,
+      "step": 800
+    },
+    {
+      "epoch": 2.556390977443609,
+      "grad_norm": 5.596978664398193,
+      "learning_rate": 4.8795180722891575e-06,
+      "loss": 0.2339,
+      "step": 850
+    },
+    {
+      "epoch": 2.706766917293233,
+      "grad_norm": 3.742239236831665,
+      "learning_rate": 4.578313253012049e-06,
+      "loss": 0.2304,
+      "step": 900
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 3.622905731201172,
+      "learning_rate": 4.27710843373494e-06,
+      "loss": 0.2317,
+      "step": 950
+    },
+    {
+      "epoch": 2.998496240601504,
+      "eval_loss": 0.189885213971138,
+      "eval_roc_auc": 0.9619820442841185,
+      "eval_runtime": 37.3548,
+      "eval_samples_per_second": 436.383,
+      "eval_steps_per_second": 1.713,
+      "step": 997
+    },
+    {
+      "epoch": 3.007518796992481,
+      "grad_norm": 3.5725858211517334,
+      "learning_rate": 3.975903614457832e-06,
+      "loss": 0.229,
+      "step": 1000
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 3.569227933883667,
+      "learning_rate": 3.6746987951807235e-06,
+      "loss": 0.1968,
+      "step": 1050
+    },
+    {
+      "epoch": 3.308270676691729,
+      "grad_norm": 5.20859432220459,
+      "learning_rate": 3.3734939759036146e-06,
+      "loss": 0.2058,
+      "step": 1100
+    },
+    {
+      "epoch": 3.4586466165413534,
+      "grad_norm": 6.604429244995117,
+      "learning_rate": 3.072289156626506e-06,
+      "loss": 0.2078,
+      "step": 1150
+    },
+    {
+      "epoch": 3.6090225563909772,
+      "grad_norm": 4.930511474609375,
+      "learning_rate": 2.771084337349398e-06,
+      "loss": 0.2053,
+      "step": 1200
+    },
+    {
+      "epoch": 3.7593984962406015,
+      "grad_norm": 3.622622489929199,
+      "learning_rate": 2.469879518072289e-06,
+      "loss": 0.2073,
+      "step": 1250
+    },
+    {
+      "epoch": 3.909774436090226,
+      "grad_norm": 4.740933895111084,
+      "learning_rate": 2.168674698795181e-06,
+      "loss": 0.2063,
+      "step": 1300
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.16450706124305725,
+      "eval_roc_auc": 0.9714800977158984,
+      "eval_runtime": 37.3773,
+      "eval_samples_per_second": 436.121,
+      "eval_steps_per_second": 1.712,
+      "step": 1330
+    },
+    {
+      "epoch": 4.06015037593985,
+      "grad_norm": 6.452109336853027,
+      "learning_rate": 1.8674698795180723e-06,
+      "loss": 0.1969,
+      "step": 1350
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 4.532094478607178,
+      "learning_rate": 1.566265060240964e-06,
+      "loss": 0.1879,
+      "step": 1400
+    },
+    {
+      "epoch": 4.360902255639098,
+      "grad_norm": 4.941292762756348,
+      "learning_rate": 1.2650602409638555e-06,
+      "loss": 0.1803,
+      "step": 1450
+    },
+    {
+      "epoch": 4.511278195488722,
+      "grad_norm": 3.9667301177978516,
+      "learning_rate": 9.638554216867472e-07,
+      "loss": 0.1821,
+      "step": 1500
+    },
+    {
+      "epoch": 4.661654135338345,
+      "grad_norm": 4.0561065673828125,
+      "learning_rate": 6.626506024096387e-07,
+      "loss": 0.1898,
+      "step": 1550
+    },
+    {
+      "epoch": 4.81203007518797,
+      "grad_norm": 4.69659423828125,
+      "learning_rate": 3.614457831325301e-07,
+      "loss": 0.1885,
+      "step": 1600
+    },
+    {
+      "epoch": 4.962406015037594,
+      "grad_norm": 4.419920444488525,
+      "learning_rate": 6.02409638554217e-08,
+      "loss": 0.1897,
+      "step": 1650
+    },
+    {
+      "epoch": 4.992481203007519,
+      "eval_loss": 0.155641108751297,
+      "eval_roc_auc": 0.9741751333869398,
+      "eval_runtime": 37.3327,
+      "eval_samples_per_second": 436.641,
+      "eval_steps_per_second": 1.714,
+      "step": 1660
+    },
+    {
+      "epoch": 4.992481203007519,
+      "step": 1660,
+      "total_flos": 4.323837898114577e+16,
+      "train_loss": 0.24427399233163122,
+      "train_runtime": 2915.5625,
+      "train_samples_per_second": 145.816,
+      "train_steps_per_second": 0.569
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 1660,
+  "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": 4.323837898114577e+16,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cbdeca5d088acc2a97c0f58c1ddd13b0642f20ee1cd04ae7049513a670cf573
+size 4859