Add application file

app.py

@@ -0,0 +1,21 @@
+import gradio as gr
+from transformers import BartTokenizer, BartForConditionalGeneration, BartConfig
+
+# Load tokenizer and model
+tokenizer = BartTokenizer.from_pretrained("facebook/bart-large-cnn")
+config = BartConfig.from_pretrained("./models/bart-summarizer/checkpoint-8000/config.json")
+model_path = "./models/bart-summarizer/checkpoint-8000/"
+model = BartForConditionalGeneration.from_pretrained(pretrained_model_name_or_path=model_path, config=config)
+
+# Define summarize function
+def summarize(text):
+    inputs = tokenizer([text], max_length=1024, return_tensors='pt', truncation=False)
+    summary_ids = model.generate(inputs['input_ids'], num_beams=4, min_length=30, max_length=128, early_stopping=True)
+    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True, clean_up_tokenization_spaces=True)
+    return summary
+
+# Create Gradio interface
+inputs = gr.Textbox(lines=10, label="Input Text")
+outputs = gr.Textbox(label="Summary")
+
+gr.Interface(summarize, inputs, outputs, title="Mail Subject Extraction", description="Get Subject from Email Content").launch()

models/bart-summarizer/checkpoint-4000/config.json

@@ -0,0 +1,70 @@
+{
+  "_name_or_path": "facebook/bart-large-cnn",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "add_final_layer_norm": false,
+  "architectures": [
+    "BartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "force_bos_token_to_be_generated": true,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "length_penalty": 2.0,
+  "max_length": 142,
+  "max_position_embeddings": 1024,
+  "min_length": 56,
+  "model_type": "bart",
+  "no_repeat_ngram_size": 3,
+  "normalize_before": false,
+  "num_beams": 4,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "prefix": " ",
+  "scale_embedding": false,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 142,
+      "min_length": 56,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 50264
+}

models/bart-summarizer/checkpoint-4000/generation_config.json

@@ -0,0 +1,16 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 2.0,
+  "max_length": 142,
+  "min_length": 56,
+  "no_repeat_ngram_size": 3,
+  "num_beams": 4,
+  "pad_token_id": 1,
+  "transformers_version": "4.39.3"
+}

models/bart-summarizer/checkpoint-4000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c75444fc36a53eaac2f4cc4194c34941c9ef7a8ea898d2858b790ce7c6f2c89c
+size 1625422896

models/bart-summarizer/checkpoint-4000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5a6b6989c79789912310711647baee8b8e1cd7545135b578f8de96fed6b0bb1
+size 3250751759

models/bart-summarizer/checkpoint-4000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7af64837f6c1802373f9a63146270fcae08cd224632d257beb2a234bda76b2b6
+size 14244

models/bart-summarizer/checkpoint-4000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a180fa12b33ed5d6c4308c080cb5d6a6cfd00e8661fe71dd10d82934574a89d4
+size 1064

models/bart-summarizer/checkpoint-4000/trainer_state.json

@@ -0,0 +1,77 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.8298755186721992,
+  "eval_steps": 500,
+  "global_step": 4000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "grad_norm": 59.103092193603516,
+      "learning_rate": 5e-05,
+      "loss": 3.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 4.993757247924805,
+      "learning_rate": 4.726477024070022e-05,
+      "loss": 1.7111,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 21.068601608276367,
+      "learning_rate": 4.452954048140044e-05,
+      "loss": 1.1954,
+      "step": 1500
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 46.41459274291992,
+      "learning_rate": 4.179431072210066e-05,
+      "loss": 0.8376,
+      "step": 2000
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.6010065078735352,
+      "learning_rate": 3.9059080962800876e-05,
+      "loss": 0.6233,
+      "step": 2500
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 57.68833541870117,
+      "learning_rate": 3.6323851203501094e-05,
+      "loss": 0.5353,
+      "step": 3000
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.1653015911579132,
+      "learning_rate": 3.358862144420131e-05,
+      "loss": 0.3499,
+      "step": 3500
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 8.111839294433594,
+      "learning_rate": 3.085339168490153e-05,
+      "loss": 0.3051,
+      "step": 4000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9640,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 4000,
+  "total_flos": 8668418408448000.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

models/bart-summarizer/checkpoint-4000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be23372fa8d5d4e5a43443bc60a636cd29afe53b764c8e48c9ac5b0cb5cd0832
+size 4920

models/bart-summarizer/checkpoint-8000/config.json

@@ -0,0 +1,70 @@
+{
+  "_name_or_path": "facebook/bart-large-cnn",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "add_final_layer_norm": false,
+  "architectures": [
+    "BartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "force_bos_token_to_be_generated": true,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "length_penalty": 2.0,
+  "max_length": 142,
+  "max_position_embeddings": 1024,
+  "min_length": 56,
+  "model_type": "bart",
+  "no_repeat_ngram_size": 3,
+  "normalize_before": false,
+  "num_beams": 4,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "prefix": " ",
+  "scale_embedding": false,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 142,
+      "min_length": 56,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 50264
+}

models/bart-summarizer/checkpoint-8000/generation_config.json

@@ -0,0 +1,16 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 2.0,
+  "max_length": 142,
+  "min_length": 56,
+  "no_repeat_ngram_size": 3,
+  "num_beams": 4,
+  "pad_token_id": 1,
+  "transformers_version": "4.39.3"
+}

models/bart-summarizer/checkpoint-8000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27c176d5344671a354389c0d9317caba88a6baa0d0db6d7dbf10c6f98319ff9b
+size 1625422896

models/bart-summarizer/checkpoint-8000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c573c59236bede059c962101b74c0a90234af3f1051904edc1924bf4eeaae862
+size 3250751759

models/bart-summarizer/checkpoint-8000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e3c987845d54ca14955bbb363cf521761b0e578be66901a29384d649b82bda
+size 14244

models/bart-summarizer/checkpoint-8000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27fd70081fb52707eb8537ab797eb0186393f9aeb921afdd7b96607215636c6b
+size 1064

models/bart-summarizer/checkpoint-8000/trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.6597510373443982,
+  "eval_steps": 500,
+  "global_step": 8000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "grad_norm": 59.103092193603516,
+      "learning_rate": 5e-05,
+      "loss": 3.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 4.993757247924805,
+      "learning_rate": 4.726477024070022e-05,
+      "loss": 1.7111,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 21.068601608276367,
+      "learning_rate": 4.452954048140044e-05,
+      "loss": 1.1954,
+      "step": 1500
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 46.41459274291992,
+      "learning_rate": 4.179431072210066e-05,
+      "loss": 0.8376,
+      "step": 2000
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.6010065078735352,
+      "learning_rate": 3.9059080962800876e-05,
+      "loss": 0.6233,
+      "step": 2500
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 57.68833541870117,
+      "learning_rate": 3.6323851203501094e-05,
+      "loss": 0.5353,
+      "step": 3000
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.1653015911579132,
+      "learning_rate": 3.358862144420131e-05,
+      "loss": 0.3499,
+      "step": 3500
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 8.111839294433594,
+      "learning_rate": 3.085339168490153e-05,
+      "loss": 0.3051,
+      "step": 4000
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.1723615527153015,
+      "learning_rate": 2.811816192560175e-05,
+      "loss": 0.206,
+      "step": 4500
+    },
+    {
+      "epoch": 1.04,
+      "grad_norm": 2.4750895500183105,
+      "learning_rate": 2.538293216630197e-05,
+      "loss": 0.1699,
+      "step": 5000
+    },
+    {
+      "epoch": 1.14,
+      "grad_norm": 1.0484932661056519,
+      "learning_rate": 2.264770240700219e-05,
+      "loss": 0.1049,
+      "step": 5500
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 0.005179296713322401,
+      "learning_rate": 1.9912472647702408e-05,
+      "loss": 0.1088,
+      "step": 6000
+    },
+    {
+      "epoch": 1.35,
+      "grad_norm": 10.785158157348633,
+      "learning_rate": 1.7177242888402626e-05,
+      "loss": 0.0777,
+      "step": 6500
+    },
+    {
+      "epoch": 1.45,
+      "grad_norm": 0.00013241704436950386,
+      "learning_rate": 1.4442013129102846e-05,
+      "loss": 0.0672,
+      "step": 7000
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 48.77959442138672,
+      "learning_rate": 1.1706783369803063e-05,
+      "loss": 0.1183,
+      "step": 7500
+    },
+    {
+      "epoch": 1.66,
+      "grad_norm": 0.09291552007198334,
+      "learning_rate": 8.971553610503283e-06,
+      "loss": 0.0474,
+      "step": 8000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9640,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 4000,
+  "total_flos": 1.7336836816896e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

models/bart-summarizer/checkpoint-8000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be23372fa8d5d4e5a43443bc60a636cd29afe53b764c8e48c9ac5b0cb5cd0832
+size 4920

requirements.txt

@@ -0,0 +1,8 @@
+accelerate==0.24.1
+gradio==3.50.2
+huggingface-hub==0.17.3
+numpy==1.24.1
+pdfplumber==0.10.3
+scipy==1.10.1
+torch
+transformers==4.35.0