Training in progress epoch 0

README.md

@@ -1,5 +1,4 @@
 ---
-license: apache-2.0
 tags:
 - generated_from_keras_callback
 model-index:
@@ -12,12 +11,12 @@ probably proofread and complete it, then remove this comment. -->
 
 # nathanReitinger/scotusPRED
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
+This model is a fine-tuned version of [pile-of-law/legalbert-large-1.7M-2](https://huggingface.co/pile-of-law/legalbert-large-1.7M-2) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.6354
-- Validation Loss: 0.6292
+- Train Loss: 0.6745
+- Validation Loss: 0.6409
 - Train Accuracy: 0.6689
-- Epoch: 2
+- Epoch: 0
 
 ## Model description
 
@@ -36,16 +35,14 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': False, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 126, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
+- optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': False, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 210, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
 - training_precision: float32
 
 ### Training results
 
 | Train Loss | Validation Loss | Train Accuracy | Epoch |
 |:----------:|:---------------:|:--------------:|:-----:|
-| 0.6428     | 0.6397          | 0.6689         | 0     |
-| 0.6408     | 0.6319          | 0.6689         | 1     |
-| 0.6354     | 0.6292          | 0.6689         | 2     |
+| 0.6745     | 0.6409          | 0.6689         | 0     |
 
 
 ### Framework versions

config.json

@@ -1,31 +1,35 @@
 {
-  "_name_or_path": "distilbert-base-uncased",
-  "activation": "gelu",
+  "_name_or_path": "pile-of-law/legalbert-large-1.7M-2",
   "architectures": [
-    "DistilBertForSequenceClassification"
+    "BertForSequenceClassification"
   ],
-  "attention_dropout": 0.1,
-  "dim": 768,
-  "dropout": 0.1,
-  "hidden_dim": 3072,
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
   "id2label": {
     "0": "Reversing",
     "1": "Affirming"
   },
   "initializer_range": 0.02,
+  "intermediate_size": 4096,
   "label2id": {
     "Affirming": 1,
     "Reversing": 0
   },
+  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "distilbert",
-  "n_heads": 12,
-  "n_layers": 6,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "pad_token_id": 0,
-  "qa_dropout": 0.1,
-  "seq_classif_dropout": 0.2,
-  "sinusoidal_pos_embds": false,
-  "tie_weights_": true,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizer",
   "transformers_version": "4.30.2",
-  "vocab_size": 30522
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 32000
 }

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dadc47f0db5098b0115879a13d212f6055337fddcf66d3cdb01c91b10015431f
-size 267951808
+oid sha256:a0ace008263ee075d88677d8f45155d2e075248d7de2fcf3bfc886d7d7dbcb15
+size 1347181616

tokenizer_config.json

@@ -1,13 +1,17 @@
 {
   "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
   "do_lower_case": true,
+  "handle_chinese_chars": true,
   "mask_token": "[MASK]",
+  "model_max_len": 512,
   "model_max_length": 512,
+  "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "strip_accents": null,
+  "strip_accents": true,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "DistilBertTokenizer",
+  "tokenizer_class": "BertTokenizer",
   "unk_token": "[UNK]"
 }