Initial upload

added_tokens.json

@@ -0,0 +1 @@
+{"[SPAN]": 30522}

conf.yml

@@ -0,0 +1,93 @@
+# Path to pretrained model or model identifier from huggingface.co/models
+model_name_or_path: "bert-base-uncased"
+
+
+train_file: "./data/train.json"
+
+dev_file: "./data/dev.json"
+
+# Pretrained config name or path if not the same as model_name
+config_name: null
+
+# Pretrained tokenizer name or path if not the same as model_name
+tokenizer_name: null
+
+# Directory to save downloaded pretrained model
+# Default to ~/.cache/huggingface/transformers
+cache_dir: null
+
+# The maximum total input sequence length.
+# Sequence longer max_seq_length will be splitted into different chunks.
+max_seq_length: 512
+
+# How many tokens should the first span have in each chunk.
+# Note that it may not be honored when the span is too long.
+doc_stride: 64
+
+# The maximum number of tokens for the hypothesis.
+# Hypotheses longer than this will be truncated.
+max_query_length: 256
+
+# Set this flag if you are using an uncased model.
+do_lower_case: true
+
+per_gpu_train_batch_size: 8
+
+per_gpu_eval_batch_size: 8
+
+learning_rate: !!float 3e-5
+
+# Number of updates steps to accumulate before performing a backward/update pass.
+gradient_accumulation_steps: 1
+
+weight_decay: 0.0
+
+adam_epsilon: !!float 1e-8
+
+max_grad_norm: 1.0
+
+num_epochs: 5.0
+
+# If set, total number of training steps to perform. Conflicts with num_epochs.
+max_steps: null
+
+# Linear warmup over warmup_steps
+warmup_steps: 200
+
+# language id of input for language-specific xlm models (see tokenization_xlm.PRETRAINED_INIT_CONFIGURATION)
+lang_id: null
+
+# Validate every n steps
+valid_steps: 3000
+
+early_stopping: true
+
+# save model every n steps
+save_steps: -1
+
+seed: 42
+
+# Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit
+fp16: false
+
+# For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3'].
+# See details at https://nvidia.github.io/apex/amp.html
+fp16_opt_level: "O1"
+
+# Make it true if you have a gpu but you don't want to use it
+no_cuda: false
+
+# Overwrite the cached training and evaluation sets
+overwrite_cache: false
+
+weight_class_probs_by_span_probs: true
+
+# class loss is multiplied by this value
+class_loss_weight: 0.1
+
+# Either of 'identification_classification' or 'classification'
+task: "identification_classification"
+
+# Whether to treat hypothesis (query) texts as a symbol instead of feeding the
+# hypothesis descriptions
+symbol_based_hypothesis: false

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "./output/best-checkpoint",
+  "architectures": [
+    "BertForIdentificationClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "class_loss_weight": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "impossible_strategy": "ignore",
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.5.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30523
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12e2951142577a28db3d3abb5fc5b7d5050a4f7503826998af749b642eb16e3a
+size 438035308

special_tokens_map.json

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

tensorboard/events.out.tfevents.1676955697.s199.uppmax.uu.se.54741.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76efb59497a331188aeb83331204c1278e942f6798842877256334f2711338af
+size 8652318

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": "/home/yiukei/.cache/huggingface/transformers/534479488c54aeaf9c3406f647aa2ec13648c06771ffe269edabebd4c412da1d.7f2721073f19841be16f41b0a70b600ca6b880c8f3df6f3535cbc704371bdfa4", "name_or_path": "bert-base-uncased"}