add files

README.md

@@ -0,0 +1,53 @@
+---
+license: mit
+tags:
+- generated_from_trainer
+datasets:
+- squad
+model-index:
+- name: microsoft_deberta-base_squad
+  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. -->
+
+# microsoft_deberta-base_squad
+
+This model is a fine-tuned version of [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base) on the squad dataset.
+
+## 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: 12
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.14.1
+- Pytorch 1.9.0
+- Datasets 1.16.1
+- Tokenizers 0.10.3

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 86.30085146641439,
+    "eval_f1": 92.68502275661561,
+    "eval_samples": 10788,
+    "train_loss": 0.6538377606771826,
+    "train_runtime": 11699.9166,
+    "train_samples": 88550,
+    "train_samples_per_second": 22.705,
+    "train_steps_per_second": 1.892
+}

config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "microsoft/deberta-base",
+  "architectures": [
+    "DebertaForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "c2p",
+    "p2c"
+  ],
+  "position_biased_input": false,
+  "relative_attention": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.14.1",
+  "type_vocab_size": 0,
+  "vocab_size": 50265
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 86.30085146641439,
+    "eval_f1": 92.68502275661561,
+    "eval_samples": 10788
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f90b764250c27c9ad1ab0edd941aeb988d77a655330921172254e171029fde
+size 554498419

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "[CLS]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "[SEP]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "[UNK]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "[SEP]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "[PAD]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "[CLS]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "[MASK]", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": {"content": "[UNK]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "[CLS]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "[SEP]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "sep_token": {"content": "[SEP]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "[CLS]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "[PAD]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "[MASK]", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "do_lower_case": false, "vocab_type": "gpt2", "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "microsoft/deberta-base", "tokenizer_class": "DebertaTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.6538377606771826,
+    "train_runtime": 11699.9166,
+    "train_samples": 88550,
+    "train_samples_per_second": 22.705,
+    "train_steps_per_second": 1.892
+}

trainer_state.json

@@ -0,0 +1,289 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 22140,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.932249322493225e-05,
+      "loss": 1.6277,
+      "step": 500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 2.86449864498645e-05,
+      "loss": 1.0907,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.7967479674796748e-05,
+      "loss": 1.0199,
+      "step": 1500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.7289972899729e-05,
+      "loss": 0.9284,
+      "step": 2000
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.6612466124661247e-05,
+      "loss": 0.9689,
+      "step": 2500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.59349593495935e-05,
+      "loss": 0.9188,
+      "step": 3000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.5257452574525747e-05,
+      "loss": 0.898,
+      "step": 3500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.4579945799457998e-05,
+      "loss": 0.902,
+      "step": 4000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.3902439024390246e-05,
+      "loss": 0.8888,
+      "step": 4500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.3224932249322494e-05,
+      "loss": 0.8703,
+      "step": 5000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 2.2547425474254745e-05,
+      "loss": 0.842,
+      "step": 5500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.186991869918699e-05,
+      "loss": 0.8403,
+      "step": 6000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.119241192411924e-05,
+      "loss": 0.8354,
+      "step": 6500
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.051490514905149e-05,
+      "loss": 0.816,
+      "step": 7000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.983739837398374e-05,
+      "loss": 0.7617,
+      "step": 7500
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.915989159891599e-05,
+      "loss": 0.5977,
+      "step": 8000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.8482384823848237e-05,
+      "loss": 0.6169,
+      "step": 8500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7804878048780488e-05,
+      "loss": 0.594,
+      "step": 9000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7127371273712736e-05,
+      "loss": 0.6107,
+      "step": 9500
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6449864498644987e-05,
+      "loss": 0.611,
+      "step": 10000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.5772357723577235e-05,
+      "loss": 0.63,
+      "step": 10500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.5094850948509487e-05,
+      "loss": 0.6039,
+      "step": 11000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4417344173441735e-05,
+      "loss": 0.6143,
+      "step": 11500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.3739837398373984e-05,
+      "loss": 0.5886,
+      "step": 12000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.3062330623306234e-05,
+      "loss": 0.6002,
+      "step": 12500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.2384823848238483e-05,
+      "loss": 0.6013,
+      "step": 13000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.1707317073170733e-05,
+      "loss": 0.5779,
+      "step": 13500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.1029810298102981e-05,
+      "loss": 0.58,
+      "step": 14000
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.035230352303523e-05,
+      "loss": 0.5889,
+      "step": 14500
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 9.674796747967479e-06,
+      "loss": 0.5174,
+      "step": 15000
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 8.997289972899728e-06,
+      "loss": 0.4231,
+      "step": 15500
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 8.319783197831978e-06,
+      "loss": 0.4261,
+      "step": 16000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 7.642276422764228e-06,
+      "loss": 0.4206,
+      "step": 16500
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 6.964769647696477e-06,
+      "loss": 0.393,
+      "step": 17000
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 6.287262872628727e-06,
+      "loss": 0.4052,
+      "step": 17500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 5.609756097560976e-06,
+      "loss": 0.4186,
+      "step": 18000
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 4.932249322493225e-06,
+      "loss": 0.4225,
+      "step": 18500
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.254742547425474e-06,
+      "loss": 0.4012,
+      "step": 19000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 3.5772357723577236e-06,
+      "loss": 0.4124,
+      "step": 19500
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 2.8997289972899733e-06,
+      "loss": 0.3895,
+      "step": 20000
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.222222222222222e-06,
+      "loss": 0.3905,
+      "step": 20500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.5447154471544717e-06,
+      "loss": 0.3781,
+      "step": 21000
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 8.672086720867209e-07,
+      "loss": 0.3939,
+      "step": 21500
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 1.8970189701897021e-07,
+      "loss": 0.408,
+      "step": 22000
+    },
+    {
+      "epoch": 3.0,
+      "step": 22140,
+      "total_flos": 6.07242628548864e+16,
+      "train_loss": 0.6538377606771826,
+      "train_runtime": 11699.9166,
+      "train_samples_per_second": 22.705,
+      "train_steps_per_second": 1.892
+    }
+  ],
+  "max_steps": 22140,
+  "num_train_epochs": 3,
+  "total_flos": 6.07242628548864e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:956f0e7ae4c82cca152ade53c0762d0e28099321cae6bd20bb0af0f6747442a4
+size 2927