Initial commit from mrm8488

README.md

@@ -0,0 +1,78 @@
+---
+language:
+- en
+license: mit
+tags:
+- generated_from_trainer
+datasets:
+- glue
+metrics:
+- accuracy
+model-index:
+- name: deberta-v3-small
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE SST2
+      type: glue
+      args: sst2
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.9403669724770642
+---
+
+<!-- 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. -->
+
+# deberta-v3-small
+
+This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the GLUE SST2 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2134
+- Accuracy: 0.9404
+
+## 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: 3e-05
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- 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 | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 0.176         | 1.0   | 4210  | 0.2134          | 0.9404   |
+| 0.1254        | 2.0   | 8420  | 0.2362          | 0.9415   |
+| 0.0957        | 3.0   | 12630 | 0.3187          | 0.9335   |
+| 0.0673        | 4.0   | 16840 | 0.3039          | 0.9266   |
+| 0.0457        | 5.0   | 21050 | 0.3521          | 0.9312   |
+
+
+### Framework versions
+
+- Transformers 4.13.0.dev0
+- Pytorch 1.10.0+cu111
+- Datasets 1.15.1
+- Tokenizers 0.10.3

added_tokens.json

@@ -0,0 +1 @@
+{"[MASK]": 128000}

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9403669724770642,
+    "eval_loss": 0.21339105069637299,
+    "eval_runtime": 1.6841,
+    "eval_samples": 872,
+    "eval_samples_per_second": 517.797,
+    "eval_steps_per_second": 64.725,
+    "train_loss": 0.10636022938685293,
+    "train_runtime": 1140.0834,
+    "train_samples": 67349,
+    "train_samples_per_second": 295.369,
+    "train_steps_per_second": 18.464
+}

config.json

@@ -0,0 +1,44 @@
+{
+  "_name_or_path": "microsoft/deberta-v3-small",
+  "architectures": [
+    "DebertaV2ForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "finetuning_task": "sst2",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "negative",
+    "1": "positive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "negative": 0,
+    "positive": 1
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.13.0.dev0",
+  "type_vocab_size": 0,
+  "vocab_size": 128100
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9403669724770642,
+    "eval_loss": 0.21339105069637299,
+    "eval_runtime": 1.6841,
+    "eval_samples": 872,
+    "eval_samples_per_second": 517.797,
+    "eval_steps_per_second": 64.725
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2b7f954e9ef086b3188c9cfd50d6d51b78cb34c0ca5012dc52ab861d654ec78
+size 567632875

runs/events.out.tfevents.1637520886.e39b553f1d72.611.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:986d61e845aede3c910460fa39bf828fe49f4758d44130be9fe8e79ba85026d0
+size 8741

runs/events.out.tfevents.1637522029.e39b553f1d72.611.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3301954d1328db322b541ce8d77d868b03f6e0698b4717090fa592fdedf148da
+size 369

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
+size 2464616

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "split_by_punct": false, "sp_model_kwargs": {}, "vocab_type": "spm", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "microsoft/deberta-v3-small", "tokenizer_class": "DebertaV2Tokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.10636022938685293,
+    "train_runtime": 1140.0834,
+    "train_samples": 67349,
+    "train_samples_per_second": 295.369,
+    "train_steps_per_second": 18.464
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_metric": 0.21339105069637299,
+  "best_model_checkpoint": "/content/deberta-v3-small/checkpoint-4210",
+  "epoch": 5.0,
+  "global_step": 21050,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.24,
+      "learning_rate": 2.8574821852731592e-05,
+      "loss": 0.2693,
+      "step": 1000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.7149643705463184e-05,
+      "loss": 0.2031,
+      "step": 2000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.5724465558194775e-05,
+      "loss": 0.1878,
+      "step": 3000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.4299287410926366e-05,
+      "loss": 0.176,
+      "step": 4000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9403669724770642,
+      "eval_loss": 0.21339105069637299,
+      "eval_runtime": 1.658,
+      "eval_samples_per_second": 525.933,
+      "eval_steps_per_second": 65.742,
+      "step": 4210
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.2874109263657958e-05,
+      "loss": 0.1353,
+      "step": 5000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.144893111638955e-05,
+      "loss": 0.1251,
+      "step": 6000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 2.002375296912114e-05,
+      "loss": 0.1309,
+      "step": 7000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.8598574821852732e-05,
+      "loss": 0.1254,
+      "step": 8000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9415137614678899,
+      "eval_loss": 0.23621003329753876,
+      "eval_runtime": 1.6921,
+      "eval_samples_per_second": 515.329,
+      "eval_steps_per_second": 64.416,
+      "step": 8420
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 1.7173396674584323e-05,
+      "loss": 0.0991,
+      "step": 9000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.5748218527315915e-05,
+      "loss": 0.087,
+      "step": 10000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.4323040380047506e-05,
+      "loss": 0.0899,
+      "step": 11000
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.2897862232779098e-05,
+      "loss": 0.0957,
+      "step": 12000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9334862385321101,
+      "eval_loss": 0.3187325596809387,
+      "eval_runtime": 1.6267,
+      "eval_samples_per_second": 536.052,
+      "eval_steps_per_second": 67.007,
+      "step": 12630
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.1472684085510689e-05,
+      "loss": 0.0747,
+      "step": 13000
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 1.004750593824228e-05,
+      "loss": 0.0609,
+      "step": 14000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 8.622327790973872e-06,
+      "loss": 0.064,
+      "step": 15000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 7.197149643705463e-06,
+      "loss": 0.0673,
+      "step": 16000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.926605504587156,
+      "eval_loss": 0.30392569303512573,
+      "eval_runtime": 1.6546,
+      "eval_samples_per_second": 527.02,
+      "eval_steps_per_second": 65.878,
+      "step": 16840
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 5.7719714964370545e-06,
+      "loss": 0.0641,
+      "step": 17000
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 4.346793349168647e-06,
+      "loss": 0.0438,
+      "step": 18000
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 2.9216152019002377e-06,
+      "loss": 0.0435,
+      "step": 19000
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 1.496437054631829e-06,
+      "loss": 0.0484,
+      "step": 20000
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 7.125890736342044e-08,
+      "loss": 0.0457,
+      "step": 21000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9311926605504587,
+      "eval_loss": 0.3521420657634735,
+      "eval_runtime": 1.6616,
+      "eval_samples_per_second": 524.793,
+      "eval_steps_per_second": 65.599,
+      "step": 21050
+    },
+    {
+      "epoch": 5.0,
+      "step": 21050,
+      "total_flos": 1.115233078066944e+16,
+      "train_loss": 0.10636022938685293,
+      "train_runtime": 1140.0834,
+      "train_samples_per_second": 295.369,
+      "train_steps_per_second": 18.464
+    }
+  ],
+  "max_steps": 21050,
+  "num_train_epochs": 5,
+  "total_flos": 1.115233078066944e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3a1522ca735076ee2b2b35ccf2f73e2ed5b250299dbaff89cedf40c2cea5430
+size 2863