Initial commit of my Gradio NLP app

app.py

@@ -0,0 +1,123 @@
+import gradio as gr
+import pandas as pd
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+from datasets import load_dataset
+from sklearn.metrics import accuracy_score, precision_recall_fscore_support, confusion_matrix
+import torch
+from sentence_transformers import SentenceTransformer
+import umap
+from sklearn.manifold import TSNE
+import matplotlib.pyplot as plt
+import seaborn as sns
+import numpy as np
+import tempfile
+from collections import Counter
+import os
+
+# Set the temporary directory for Gradio to use
+os.environ['GRADIO_TEMP_DIR'] = '/home/marwen/gradio_tmp'
+
+# Load the models and their tokenizers
+model_paths = {
+    "roberta-base-offensive": "./models/roberta-base-offensive",
+    "distilbert-base-uncased-offensive": "./models/distilbert-base-uncased-offensive",
+    "bert-offensive": "./models/bert-offensive",
+    "deberta-offensive": "./models/deberta-offensive"
+}
+
+models = {name: AutoModelForSequenceClassification.from_pretrained(path) for name, path in model_paths.items()}
+tokenizers = {name: AutoTokenizer.from_pretrained(path) for name, path in model_paths.items()}
+
+# Load the dataset
+dataset = load_dataset("tweet_eval", "offensive")
+
+# Initialize Sentence Transformer for embedding generation
+model_embedding = SentenceTransformer('all-MiniLM-L6-v2')
+
+def encode(texts, tokenizer):
+    return tokenizer(texts, padding="max_length", truncation=True, max_length=128, return_tensors="pt")
+
+def predict(model, inputs):
+    model.eval()
+    with torch.no_grad():
+        outputs = model(**inputs)
+        preds = outputs.logits.argmax(-1).cpu().numpy()
+    return preds
+
+def calculate_metrics(labels, preds):
+    accuracy = accuracy_score(labels, preds)
+    precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average='binary')
+    conf_matrix = confusion_matrix(labels, preds)
+    return accuracy, precision, recall, f1, conf_matrix
+
+def generate_confusion_matrix(conf_matrix, model_name):
+    plt.figure(figsize=(5, 4))
+    sns.heatmap(conf_matrix, annot=True, fmt="d")
+    plt.title(f'Confusion Matrix: {model_name}')
+    plt.ylabel('Actual')
+    plt.xlabel('Predicted')
+    plt.tight_layout()
+    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.png')
+    plt.savefig(temp_file.name)
+    plt.close()
+    return temp_file.name
+
+def generate_embeddings_and_plot(categories):
+    all_texts = sum(categories.values(), [])
+    embeddings = model_embedding.encode(all_texts)
+    umap_reducer = umap.UMAP(n_neighbors=15, n_components=2, metric='cosine')
+    umap_embeddings = umap_reducer.fit_transform(embeddings)
+    tsne_embeddings = TSNE(n_components=2, perplexity=30).fit_transform(embeddings)
+
+    def plot_embeddings(embeddings, title, file_suffix):
+        plt.figure(figsize=(10, 8))
+        colors = {"correct_both": "green", "incorrect_both": "red", "correct_model1_only": "blue", "correct_model2_only": "orange"}
+        for category, color in colors.items():
+            indices = [i for i, text in enumerate(all_texts) if text in categories[category]]
+            plt.scatter(embeddings[indices, 0], embeddings[indices, 1], label=category, color=color, alpha=0.6)
+        plt.legend()
+        plt.title(title)
+        plt.xlabel('Component 1')
+        plt.ylabel('Component 2')
+        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=f'_{file_suffix}.png')
+        plt.savefig(temp_file.name)
+        plt.close()
+        return temp_file.name
+
+    umap_plot_path = plot_embeddings(umap_embeddings, "UMAP Projection of Text Categories", "umap")
+    tsne_plot_path = plot_embeddings(tsne_embeddings, "t-SNE Projection of Text Categories", "tsne")
+    return umap_plot_path, tsne_plot_path
+
+def setup_gradio_interface():
+    with gr.Blocks() as demo:
+        gr.Markdown("## Model Comparison and Text Analysis")
+        with gr.Row():
+            model1_input = gr.Dropdown(list(model_paths.keys()), label="Model 1")
+            model2_input = gr.Dropdown(list(model_paths.keys()), label="Model 2")
+        submit_button = gr.Button("Compare")
+        
+        metrics_output = gr.Dataframe()
+        
+        with gr.Row():
+            model1_cm_output = gr.Image(label="Confusion Matrix for Model 1")
+            model2_cm_output = gr.Image(label="Confusion Matrix for Model 2")
+        
+        with gr.Row():
+            umap_visualization_output = gr.Image(label="UMAP Text Categorization Visualization")
+            tsne_visualization_output = gr.Image(label="t-SNE Text Categorization Visualization")
+
+        def update_interface(model1, model2):
+            metrics_df, conf_matrix1, conf_matrix2 = compare_models(model1, model2)
+            umap_plot_path, tsne_plot_path = generate_embeddings_and_plot(metrics_df)
+            return metrics_df, conf_matrix1, conf_matrix2, umap_plot_path, tsne_plot_path
+
+        submit_button.click(
+            update_interface,
+            inputs=[model1_input, model2_input],
+            outputs=[metrics_output, model1_cm_output, model2_cm_output, umap_visualization_output, tsne_visualization_output]
+        )
+        
+    return demo
+
+demo = setup_gradio_interface()
+demo.launch(share=True)

models/UBC-NLP/ARBERT/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "UBC-NLP/ARBERT",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 100000
+}

models/UBC-NLP/ARBERT/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5721a08143fe631ff017403f0a526b5d570444ebab9af55869f50815112efd5
+size 651395072

models/UBC-NLP/ARBERT/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

models/UBC-NLP/ARBERT/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

models/UBC-NLP/MARBERT/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "UBC-NLP/MARBERT",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 100000
+}

models/UBC-NLP/MARBERT/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:874e85391a27658550ccfde41216bf0d1bf732472e4461d8ba7876a4dcd1b2c6
+size 651395072

models/UBC-NLP/MARBERT/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

models/UBC-NLP/MARBERT/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-1000/config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

models/bert-offensive/checkpoint-1000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd658ae002182ad99ffbb9982d5d85515c0c49949eac2d6d4eb86c479d67955b
+size 437958648

models/bert-offensive/checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22c5eb3685348006d842a33f8c23a222816e6efa7e3dbf1212516a44add594c0
+size 876038394

models/bert-offensive/checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89ed3ecad87f77a0ed60753e4ffd969dbef607419cab5d7c3471fc6b16360766
+size 14244

models/bert-offensive/checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cdc454fd78e2054130f997d6bf748bc81368c748607a351ea7de15a408d26e6
+size 1064

models/bert-offensive/checkpoint-1000/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-1000/trainer_state.json

@@ -0,0 +1,35 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6711409395973155,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.34,
+      "grad_norm": 3.309788227081299,
+      "learning_rate": 4.440715883668904e-05,
+      "loss": 0.5291,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 8.38113784790039,
+      "learning_rate": 3.881431767337807e-05,
+      "loss": 0.4834,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4470,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 423493688042880.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

models/bert-offensive/checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff52faf0d3a27635d14f511f07bc80871b57c3a7826358862dda78696619e8f5
+size 4856

models/bert-offensive/checkpoint-1500/config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

models/bert-offensive/checkpoint-1500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b13c8461e29b1c84d3992da1258c4e67c9df1faa1b7d2a6d3866c8b4ab31264e
+size 437958648

models/bert-offensive/checkpoint-1500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a5c46aedd4909a23c89b0d3396dc7938f781b6289de190c8bb1fa275715998e
+size 876038394

models/bert-offensive/checkpoint-1500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a5e93a639b18c5080b3206380db8bdf3c9bf762e376ba68470d9bd63251d88
+size 14244

models/bert-offensive/checkpoint-1500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bd70d973811a00199980200e5162073c3f4bfdfc513c94464ee46ef32af9473
+size 1064

models/bert-offensive/checkpoint-1500/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-1500/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-1500/trainer_state.json

@@ -0,0 +1,54 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0067114093959733,
+  "eval_steps": 500,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.34,
+      "grad_norm": 3.309788227081299,
+      "learning_rate": 4.440715883668904e-05,
+      "loss": 0.5291,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 8.38113784790039,
+      "learning_rate": 3.881431767337807e-05,
+      "loss": 0.4834,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8244186046511628,
+      "eval_f1": 0.700990099009901,
+      "eval_loss": 0.40849873423576355,
+      "eval_precision": 0.6679245283018868,
+      "eval_recall": 0.7375,
+      "eval_runtime": 1.415,
+      "eval_samples_per_second": 607.771,
+      "eval_steps_per_second": 76.325,
+      "step": 1490
+    },
+    {
+      "epoch": 1.01,
+      "grad_norm": 4.857753753662109,
+      "learning_rate": 3.3221476510067115e-05,
+      "loss": 0.4743,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4470,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 635261087615520.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

models/bert-offensive/checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff52faf0d3a27635d14f511f07bc80871b57c3a7826358862dda78696619e8f5
+size 4856

models/bert-offensive/checkpoint-2000/config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

models/bert-offensive/checkpoint-2000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d84caca90c708c6d9b08f796f62f3fe3770d1ebbe5dc4708d787c870652e882a
+size 437958648

models/bert-offensive/checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f977dfb26bf16852ccff9864f65c99dfca153be14bcd7280af5f7a05a0d808
+size 876038394

models/bert-offensive/checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e57db07813490de347f6b29da476f84f897496ba4cc251212c6d017a619dce80
+size 14244

models/bert-offensive/checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd68bf4dff9924c9178e7ecdd91ec247aec55edc72b3ca0231603fd33cef86d3
+size 1064

models/bert-offensive/checkpoint-2000/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

models/bert-offensive/checkpoint-2000/trainer_state.json

@@ -0,0 +1,61 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.342281879194631,
+  "eval_steps": 500,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.34,
+      "grad_norm": 3.309788227081299,
+      "learning_rate": 4.440715883668904e-05,
+      "loss": 0.5291,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 8.38113784790039,
+      "learning_rate": 3.881431767337807e-05,
+      "loss": 0.4834,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8244186046511628,
+      "eval_f1": 0.700990099009901,
+      "eval_loss": 0.40849873423576355,
+      "eval_precision": 0.6679245283018868,
+      "eval_recall": 0.7375,
+      "eval_runtime": 1.415,
+      "eval_samples_per_second": 607.771,
+      "eval_steps_per_second": 76.325,
+      "step": 1490
+    },
+    {
+      "epoch": 1.01,
+      "grad_norm": 4.857753753662109,
+      "learning_rate": 3.3221476510067115e-05,
+      "loss": 0.4743,
+      "step": 1500
+    },
+    {
+      "epoch": 1.34,
+      "grad_norm": 20.212223052978516,
+      "learning_rate": 2.7628635346756153e-05,
+      "loss": 0.3753,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4470,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 846855820558080.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

models/bert-offensive/checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff52faf0d3a27635d14f511f07bc80871b57c3a7826358862dda78696619e8f5
+size 4856

models/bert-offensive/checkpoint-2500/config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

models/bert-offensive/checkpoint-2500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f58457eeb8339abe1f0ee26937443a95a8a25aae3c4fe5f569327ab3fd87f48c
+size 437958648

models/bert-offensive/checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77406637943250708e3999e599a8b10ef39a486f27cfa07a01bbff9f9a76e71b
+size 876038394

models/bert-offensive/checkpoint-2500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ccff0626d2d12f751c557becd9d0ba8acfabed972b2c6065108c51d3e8d4d12
+size 14244