FEAT: New version of model trained as sequence classification

app.py

@@ -22,7 +22,7 @@ def get_model(_model_path):
             _model_package = meta_info["kwargs"].get("model_package", "transformers")
             _model_class = meta_info["kwargs"].get("model_class", "BertForNextSentencePrediction")
     else:
-        raise FileNotFoundError("Tokenizer is provided without meta-info.json. Cannot interfere proper configuration!")
+        raise FileNotFoundError("Model is provided without meta-info.json. Cannot interfere proper configuration!")
 
     model_class = get_class(_model_package, _model_class)
     _model = model_class.from_pretrained(_model_path)
@@ -32,8 +32,8 @@ def get_model(_model_path):
 
 def get_tokenizer(tokenizer_path):
     print(f"Getting tokenizer at {tokenizer_path}")
-    from transformers import BertTokenizer
-    tokenizer = BertTokenizer.from_pretrained(tokenizer_path)
+    from transformers import AutoTokenizer
+    tokenizer = AutoTokenizer.from_pretrained(tokenizer_path)
     if os.path.isfile(os.path.join(tokenizer_path, "meta-info.json")):
         with open(os.path.join(tokenizer_path, "meta-info.json"), "r") as f:
             meta_info = json.load(f)
@@ -44,8 +44,7 @@ def get_tokenizer(tokenizer_path):
 
     if special_token != " ":
         tokenizer.add_special_tokens({"additional_special_tokens": [special_token]})
-    # print(special_token)
-    # print(tokenizer_args)
+
     _inference_tokenizer = NextSentencePredictionTokenizer(tokenizer, **tokenizer_args)
     return _inference_tokenizer
 
@@ -108,7 +107,10 @@ option = st.selectbox("Choose type of input:",
                        "02 - JSON (aggregated)",
                        "03 - JSON (example CA-OOD)",
                        "04 - JSON (example Elysai)",
-                       "05 - Diagnostic mode"])
+                       "05 - Diagnostic mode",
+                       "06 - JSON (example Elysai - large)"])
+
+progres_bar = st.progress(0.0, text="Inference")
 
 with st.form("input_text"):
     if "01" in option:
@@ -129,10 +131,9 @@ with st.form("input_text"):
             ax.pie([prop_follow, prop_not_follow], labels=["Probability - Follow", "Probability - Not Follow"],
                    autopct='%1.1f%%')
             st.pyplot(fig)
-    elif "02" in option or "03" in option or "04" in option:
-        from data.example_data import ca_ood, elysai
+    elif "02" in option or "03" in option or "04" in option or "06" in option:
+        from data.example_data import ca_ood, elysai, elysai_large
 
-        choices = [ca_ood, elysai]
         option: str
         # > Python 3.10
         # match option.split("-")[0].strip():
@@ -143,12 +144,14 @@ with st.form("input_text"):
         #     case _:
         #         text = ""
         option = option.split("-")[0].strip()
+        text = ""
         if option == "03":
-            text = json.dumps(choices[0])
+            text = json.dumps(ca_ood)
         elif option == "04":
-            text = json.dumps(choices[1])
-        else:
-            test = ""
+            text = json.dumps(elysai)
+        elif option == "06":
+            text = json.dumps(elysai_large)
+
         context = st.text_area("Insert JSON here:", value=str(text))
 
         if "{" in context:
@@ -158,7 +161,8 @@ with st.form("input_text"):
 
         submitted = st.form_submit_button("Submit")
         if submitted:
-            for datapoint in data_for_evaluation:
+            for idx, datapoint in enumerate(data_for_evaluation):
+                progres_bar.progress(idx/len(data_for_evaluation), text="Inference")
                 c, s, human_label = datapoint
                 input_tensor = inference_tokenizer.get_item(context=c, actual_sentence=s)
                 output_model = model(**input_tensor.data).logits

inference_tokenizer.py

@@ -1,7 +1,6 @@
 import torch
 from typing import Dict, List
 
-
 class NextSentencePredictionTokenizer:
 
     def __init__(self, _tokenizer, **_tokenizer_args):

model/3c090bb39725194fae09a603caac8c3d9014df49/awscliv2.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4962990f24634814b6d4834a5c105a524c8895fca478a8fc17f7cc7e6191fa4
+size 57717779

model/3c090bb39725194fae09a603caac8c3d9014df49/onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1e7543fd13846a928c1309459cd3a068dc82221a4d22ca647e054d658ea1063
+size 433513952

model/4a70ad1033ceec48447d5319d0863d442e976823/info.json

@@ -0,0 +1,4 @@
+{
+  "model": "BERT-NSP-v7",
+  "description": "Model trained on full version of DailyDialogue and CommonDialogues + down=sampled version of SODA and AirDialogue. Using [unused1] token to divide sentences in context. Improved training arguments (warmup, smaller learning rate). Using frozen test set to better compare models and therefore trained longer time (about 60 epochs). The model also have bigger classification head (from one layer liner as classical). Added method for **Data Augmentation**. More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/b40zgvoo/overview?workspace=user-petr-lorenc"
+}

model/4a70ad1033ceec48447d5319d0863d442e976823/meta-info.json

@@ -0,0 +1,56 @@
+{
+  "args": [],
+  "kwargs": {
+    "model_package": "models",
+    "model_class": "OwnBertForNextSentencePrediction",
+    "data_root": "/home/lorenpe2/project/data",
+    "data_sources": [
+      [
+        "COMMON_DIALOGUES",
+        "common_dialogues/train.json",
+        "common_dialogues/valid_frozen.json",
+        "common_dialogues/test_frozen.json"
+      ],
+      [
+        "DAILY_DIALOGUES",
+        "daily_dialogues/dialogues_text.train.txt",
+        "daily_dialogues/dev_frozen.json",
+        "daily_dialogues/test_frozen.json"
+      ],
+      [
+        "DAILY_DIALOGUES",
+        "air_dialogue/subsampled_train.txt",
+        "air_dialogue/subsampled_validation_frozen.json",
+        "air_dialogue/subsampled_test_frozen.json"
+      ],
+      [
+        "DAILY_DIALOGUES",
+        "soda/subsampled_train.txt",
+        "soda/subsampled_validation_frozen.json",
+        "soda/subsampled_test_frozen.json"
+      ]
+    ],
+    "pretrained_model": "bert-base-uncased",
+    "tokenizer": "bert-base-uncased",
+    "approach": "UNIQUE_RANDOM_CONTEXT",
+    "data_augmentation": [
+      "ADD_PARTLY_BROKEN_CONTEXT",
+      "ADD_SMALLER_CONTEXT"
+    ],
+    "special_token": "[unused1]",
+    "learning_rate": 5e-07,
+    "warmup_ratio": 0.1,
+    "freeze_prefinetuning": true,
+    "prefinenuting_epoch": 10,
+    "finetuning_epochs": 75
+  },
+  "tokenizer_args": {
+    "padding": "max_length",
+    "max_length_ctx": 32,
+    "max_length_res": 8,
+    "truncation": "only_first",
+    "return_tensors": "np",
+    "is_split_into_words": true,
+    "special_token": "[unused1]"
+  }
+}

model/{6a62f122a90e090b285f0344a1d79e753f2000bb → 4a70ad1033ceec48447d5319d0863d442e976823}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb065d01ae7e4e255d81ca3fdafadb21c42daea7cafdaaac3c21923f11827641
+oid sha256:8a0200d7532286b0e8aae550933d8a083274c7d8bcba41bd1ac989f9efb1bb1d
 size 438871109

model/{e09d71f55f4b6fc20135f856bf029322a3265d8d → 4a70ad1033ceec48447d5319d0863d442e976823}/special_tokens_map.json

@@ -1,4 +1,7 @@
 {
+  "additional_special_tokens": [
+    "[unused1]"
+  ],
   "cls_token": "[CLS]",
   "mask_token": "[MASK]",
   "pad_token": "[PAD]",

model/4a70ad1033ceec48447d5319d0863d442e976823/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

model/{6a62f122a90e090b285f0344a1d79e753f2000bb → 4a70ad1033ceec48447d5319d0863d442e976823}/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0032b55f2a68888f89b97e84498c69d4a7a65403f1a209c41da390a1762f76fc
+oid sha256:fb391844e5d1b871e851254cc8388e4803682dd147af1fdb5067fa64fbe530aa
 size 3195

model/6a62f122a90e090b285f0344a1d79e753f2000bb/info.json

@@ -1,4 +0,0 @@
-{
-  "model": "BERT-NSP-v6",
-  "description": "Model trained on DailyDialogue and CommonDialogues. Using [unused1] token to divide sentences in context. Improved training arguments (warmup, smaller learning rate). Using frozen test set to better compare models and therefore trained longer time (about 60 epochs). The model also have bigger classification head (from one layer liner as classical). Added method for **Data Augmentation**. More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/xvzhhw2r/overview?workspace=user-petr-lorenc"
-}

model/6a62f122a90e090b285f0344a1d79e753f2000bb/meta-info.json

@@ -1 +0,0 @@
-{"args": [], "kwargs": {"model_package": "models", "model_class": "OwnBertForNextSentencePrediction", "data_root": "/home/lorenpe2/project/data", "data_sources": [["COMMON_DIALOGUES", "common_dialogues/train.json", "common_dialogues/valid_frozen.json", "common_dialogues/test_frozen.json"], ["DAILY_DIALOGUES", "daily_dialogues/dialogues_text.train.txt", "daily_dialogues/dev_frozen.json", "daily_dialogues/test_frozen.json"]], "pretrained_model": "bert-base-uncased", "tokenizer": "bert-base-uncased", "approach": "IGNORE_DUPLICITIES", "data_augmentation": ["ADD_PARTLY_BROKEN_CONTEXT"], "special_token": "[unused1]", "learning_rate": 5e-07, "warmup_ratio": 0.1, "freeze_prefinetuning": true, "prefinenuting_epoch": 10, "finetuning_epochs": 75}, "tokenizer_args": {"padding": "max_length", "max_length_ctx": 256, "max_length_res": 40, "truncation": "only_first", "return_tensors": "np", "is_split_into_words": true, "approach": "IGNORE_DUPLICITIES", "special_token": "[unused1]"}}

model/6a62f122a90e090b285f0344a1d79e753f2000bb/special_tokens_map.json

@@ -1 +0,0 @@
-{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "additional_special_tokens": ["[unused1]"]}

model/6a62f122a90e090b285f0344a1d79e753f2000bb/tokenizer_config.json

@@ -1 +0,0 @@
-{"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, "name_or_path": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}

model/berttokenizer.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d253965be810ac94eed5cf080d45c958e282f0450d64dff77803b272e145d8e
+size 320047

model/berttokenizer/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]",
+  "additional_special_tokens": ["[unused1]", "[unused2]", "[unused3]", "[unused4]", "[unused5]", "[unused6]", "[unused7]", "[unused8]", "[unused9]", "[unused10]"]
+}

model/{e09d71f55f4b6fc20135f856bf029322a3265d8d → berttokenizer}/tokenizer.json

@@ -5,48 +5,48 @@
   "added_tokens": [
     {
       "id": 0,
+      "special": true,
       "content": "[PAD]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
-      "special": true
+      "normalized": false
     },
     {
       "id": 100,
+      "special": true,
       "content": "[UNK]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
-      "special": true
+      "normalized": false
     },
     {
       "id": 101,
+      "special": true,
       "content": "[CLS]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
-      "special": true
+      "normalized": false
     },
     {
       "id": 102,
+      "special": true,
       "content": "[SEP]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
-      "special": true
+      "normalized": false
     },
     {
       "id": 103,
+      "special": true,
       "content": "[MASK]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
-      "special": true
+      "normalized": false
     }
   ],
   "normalizer": {

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → berttokenizer}/tokenizer_config.json

@@ -3,6 +3,7 @@
   "do_lower_case": true,
   "mask_token": "[MASK]",
   "model_max_length": 512,
+  "name_or_path": "bert-base-uncased",
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
   "special_tokens_map_file": null,

model/c3c3bdb7ad80396e69de171995e2038f900940c8/info.json

@@ -1,4 +0,0 @@
-{
-  "model": "BERT-NSP-v1",
-  "description": "Model trained on DailyDialogue. Context is taken as is - no separation of individual turns. More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/dm6ab7ma/logs?workspace=user-petr-lorenc"
-}

model/c3c3bdb7ad80396e69de171995e2038f900940c8/meta-info.json

@@ -1 +0,0 @@
-{"args": [], "kwargs": {"data_root": "/home/lorenpe2/project/data", "data_sources": [["COMMON_DIALOGUES", "common_dialogues/train.json", "common_dialogues/valid.json", "common_dialogues/test.json"]], "pretrained_model": "bert-base-uncased", "tokenizer": "bert-base-uncased", "naive_approach": true, "special_token": " ", "learning_rate": 5e-05}, "tokenizer_args": {"padding": "max_length", "max_length_ctx": 256, "max_length_res": 64, "truncation": "only_first", "return_tensors": "np", "is_split_into_words": true, "naive_approach": true, "special_token": " "}}

model/d1dd8365cbf16ff423f537e2291c61a91c717ed1/onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a65ee19216a6cfe592c1b5d7b35dabe9182a60d9bf7dbe415a7ad491b6c64733
+size 438044019

model/e09d71f55f4b6fc20135f856bf029322a3265d8d/info.json

@@ -1,4 +0,0 @@
-{
-  "model": "BERT-NSP-v2",
-  "description": "Model trained on DailyDialogue and CommonDialogues. Using [unused1] token to divide sentences in context. More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/ll69cliu/logs?workspace=user-petr-lorenc"
-}

model/e09d71f55f4b6fc20135f856bf029322a3265d8d/meta-info.json

@@ -1 +0,0 @@
-{"args": [], "kwargs": {"data_root": "/home/lorenpe2/project/data", "data_sources": [["COMMON_DIALOGUES", "common_dialogues/train.json", "common_dialogues/valid.json", "common_dialogues/test.json"], ["DAILY_DIALOGUES", "daily_dialogues/dialogues_text.train.txt", "daily_dialogues/dialogues_text.dev.txt", "daily_dialogues/dialogues_text.test.txt"]], "pretrained_model": "bert-base-uncased", "tokenizer": "bert-base-uncased", "naive_approach": true, "special_token": "[unused1]", "learning_rate": 5e-05}, "tokenizer_args": {"padding": "max_length", "max_length_ctx": 256, "max_length_res": 64, "truncation": "only_first", "return_tensors": "np", "is_split_into_words": true, "naive_approach": true, "special_token": "[unused1]"}}

model/e09d71f55f4b6fc20135f856bf029322a3265d8d/optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ab19babe7ae39d1b6746d1dec58ab7758b0bacc33346b66e9c7da94419bebf96
-size 124944384

model/f1f881389fb38108e623689999ceaaaf398c5e92/info.json

@@ -1,4 +0,0 @@
-{
-  "model": "BERT-NSP-v5",
-  "description": "Model trained on DailyDialogue and CommonDialogues. Using [unused1] token to divide sentences in context. Improved training arguments (warmup, smaller learning rate). Using frozen test set to better compare models and therefore trained longer time (about 60 epochs). The model also have bigger classification head (from one layer liner as classical). More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/vzpwetvm/overview?workspace=user-petr-lorenc"
-}

model/f1f881389fb38108e623689999ceaaaf398c5e92/meta-info.json

@@ -1 +0,0 @@
-{"args": [], "kwargs": {"model_package": "models", "model_class": "OwnBertForNextSentencePrediction", "data_root": "/home/lorenpe2/project/data", "data_sources": [["COMMON_DIALOGUES", "common_dialogues/train.json", "common_dialogues/valid_frozen.json", "common_dialogues/test_frozen.json"], ["DAILY_DIALOGUES", "daily_dialogues/dialogues_text.train.txt", "daily_dialogues/dev_frozen.json", "daily_dialogues/test_frozen.json"]], "pretrained_model": "bert-base-uncased", "tokenizer": "bert-base-uncased", "approach": "IGNORE_DUPLICITIES", "special_token": "[unused1]", "learning_rate": 5e-07, "warmup_ratio": 0.1, "freeze_prefinetuning": true, "prefinenuting_epoch": 10, "finetuning_epochs": 75}, "tokenizer_args": {"padding": "max_length", "max_length_ctx": 256, "max_length_res": 40, "truncation": "only_first", "return_tensors": "np", "is_split_into_words": true, "approach": "IGNORE_DUPLICITIES", "special_token": "[unused1]"}}

model/f1f881389fb38108e623689999ceaaaf398c5e92/special_tokens_map.json

@@ -1 +0,0 @@
-{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "additional_special_tokens": ["[unused1]"]}

model/f1f881389fb38108e623689999ceaaaf398c5e92/tokenizer_config.json

@@ -1 +0,0 @@
-{"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, "name_or_path": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}

model/{e09d71f55f4b6fc20135f856bf029322a3265d8d → new_version_dummy}/config.json

@@ -1,7 +1,7 @@
 {
   "_name_or_path": "bert-base-uncased",
   "architectures": [
-    "BertForNextSentencePrediction"
+    "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
@@ -18,8 +18,9 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.17.0",
+  "transformers_version": "4.30.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

model/new_version_dummy/onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adf2c61611e66c3efcca5e6d866b354596fff14c0f5d49ef4ff73bcbc77a20bc
+size 438201824

model/{e09d71f55f4b6fc20135f856bf029322a3265d8d → new_version_dummy}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f791a2952d2707e82b275b14738b0fcd52c56b9a6acd597f4480829737d4368
-size 438022005
+oid sha256:04a843b2644531a9f6ac15659f22d89bc288c17886e9c6d2561431a7a0add441
+size 438007925

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → new_version_dummy}/special_tokens_map.json

@@ -1,4 +1,7 @@
 {
+  "additional_special_tokens": [
+    "[unused1]"
+  ],
   "cls_token": "[CLS]",
   "mask_token": "[MASK]",
   "pad_token": "[PAD]",

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → new_version_dummy}/tokenizer.json

@@ -12,6 +12,15 @@
       "normalized": false,
       "special": true
     },
+    {
+      "id": 2,
+      "content": "[unused1]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
     {
       "id": 100,
       "content": "[UNK]",

model/{e09d71f55f4b6fc20135f856bf029322a3265d8d → new_version_dummy}/tokenizer_config.json

@@ -1,11 +1,11 @@
 {
+  "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
   "do_lower_case": true,
   "mask_token": "[MASK]",
   "model_max_length": 512,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "special_tokens_map_file": null,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "BertTokenizer",

model/{f1f881389fb38108e623689999ceaaaf398c5e92 → new_version_dummy}/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49a0183a62c25be44cbf2f333cf9224204e7fbfe84cd9054d94775d61daa9774
-size 3195
+oid sha256:a5e18fad4f08f10e21116128e5ffdbc3b9da56804fae548529abfcafe8814d94
+size 3899

model/{f1f881389fb38108e623689999ceaaaf398c5e92 → new_version_v1}/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "bert-base-uncased",
+  "_name_or_path": "/home/lorenpe2/project/hf_models/bert-base-uncased",
   "architectures": [
     "OwnBertForNextSentencePrediction"
   ],
@@ -19,7 +19,7 @@
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.17.0",
+  "transformers_version": "4.30.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

model/new_version_v1/onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a4908d96ecead256b5c4f921a6ae08fe156bdb3a11aceb5aa7b8677dac9322e
+size 438174352

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → new_version_v1}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aec095ee8fc2c88ca0460f59e19df6b38c5c91d38a3ab04928ce7eb996c0d62a
-size 438022005
+oid sha256:7ed3691d1c07d77a0a6796fb4a5f61a2e774849cb7489c97a183097cb4f693aa
+size 438856837

model/new_version_v1/special_tokens_map.json

@@ -0,0 +1,10 @@
+{
+  "additional_special_tokens": [
+    "[unused1]"
+  ],
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

model/new_version_v1/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → new_version_v1}/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9965b76852aa7c689ed048738e3db2f0b386154cddbfb42d9da7a064a9f9231
-size 3195
+oid sha256:bed99c1c5b736531dce1c0822e055bb8ad10a1634b9235d9a35e7d6aa4356d58
+size 4091

model/{c3c3bdb7ad80396e69de171995e2038f900940c8 → new_version_v2}/config.json

@@ -1,7 +1,7 @@
 {
-  "_name_or_path": "bert-base-uncased",
+  "_name_or_path": "/home/lorenpe2/project/hf_models/bert-base-uncased",
   "architectures": [
-    "BertForNextSentencePrediction"
+    "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
@@ -18,8 +18,9 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.17.0",
+  "transformers_version": "4.30.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

model/new_version_v2/info.json

@@ -0,0 +1,4 @@
+{
+  "model": "BERT-SEQUENCE-CLASSIFICATION",
+  "description": "Model trained on subset of DailyDialogue, CommonDialogues, ChitChatDataset, AirDialogue and SODA. Using [unused1] token to divide sentences in context. More info can be found at https://wandb.ai/alquist/next-sentence-prediction/runs/wfsx1sga/overview?workspace=user-petr-lorenc"
+}

model/new_version_v2/meta-info.json

@@ -0,0 +1,28 @@
+{
+  "args": [],
+  "kwargs": {
+    "model_package": "transformers",
+    "model_class": "AutoModelForSequenceClassification",
+    "data_root": "/home/lorenpe2/project/data",
+    "data_sources": [],
+    "pretrained_model": "bert-base-uncased",
+    "tokenizer": "bert-base-uncased",
+    "approach": "IGNORE_DUPLICITIES",
+    "special_token": "[unused1]",
+    "learning_rate": 5e-07,
+    "warmup_ratio": 0.1,
+    "freeze_prefinetuning": true,
+    "prefinenuting_epoch": 10,
+    "finetuning_epochs": 75
+  },
+  "tokenizer_args": {
+    "padding": "max_length",
+    "max_length_ctx": 256,
+    "max_length_res": 40,
+    "truncation": "only_first",
+    "return_tensors": "np",
+    "is_split_into_words": true,
+    "approach": "IGNORE_DUPLICITIES",
+    "special_token": "[unused1]"
+  }
+}