deployment files added

README.md

@@ -6,6 +6,7 @@ colorTo: blue
 sdk: docker
 pinned: false
 license: mit
+app_port:8081
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,52 @@
+import gradio as gr
+
+from transformers import LineByLineTextDataset
+from transformers import DataCollatorForLanguageModeling
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
+from transformers import Trainer, TrainingArguments
+
+def load_model(model_path):
+    model = GPT2LMHeadModel.from_pretrained(model_path)
+    return model
+
+
+def load_tokenizer(tokenizer_path):
+    tokenizer = GPT2Tokenizer.from_pretrained(tokenizer_path)
+    return tokenizer
+
+
+def generate_text(sequence, max_new_tokens):
+    ids = tokenizer.encode(f'{sequence}', return_tensors='pt')
+    input_length = ids.size(1)
+    max_length = input_length + max_new_tokens
+    final_outputs = model.generate(
+        ids,
+        do_sample=True,
+        max_length=max_length,
+        pad_token_id=model.config.eos_token_id
+    )
+    return tokenizer.decode(final_outputs[0], skip_special_tokens=True)
+
+
+
+
+model_path = r'C:\Users\Suraj Singh\aiml_gr16\checkpoint'
+model = load_model(model_path)
+tokenizer = load_tokenizer(model_path)
+
+
+def generate_text(sequence, max_new_tokens):
+    ids = tokenizer.encode(f'{sequence}', return_tensors='pt')
+    input_length = ids.size(1)
+    max_length = input_length + max_new_tokens
+    final_outputs = model.generate(
+        ids,
+        do_sample=True,
+        max_length=max_length,
+        pad_token_id=model.config.eos_token_id
+    )
+    return tokenizer.decode(final_outputs[0], skip_special_tokens=True)
+
+iface = gr.Interface(fn=generate_text, inputs="text", outputs="text")
+
+iface.launch()

checkpoint/config.json

@@ -0,0 +1,39 @@
+{
+    "_name_or_path": "gpt2",
+    "activation_function": "gelu_new",
+    "architectures": [
+      "GPT2LMHeadModel"
+    ],
+    "attn_pdrop": 0.1,
+    "bos_token_id": 50256,
+    "embd_pdrop": 0.1,
+    "eos_token_id": 50256,
+    "initializer_range": 0.02,
+    "layer_norm_epsilon": 1e-05,
+    "model_type": "gpt2",
+    "n_ctx": 1024,
+    "n_embd": 768,
+    "n_head": 12,
+    "n_inner": null,
+    "n_layer": 12,
+    "n_positions": 1024,
+    "reorder_and_upcast_attn": false,
+    "resid_pdrop": 0.1,
+    "scale_attn_by_inverse_layer_idx": false,
+    "scale_attn_weights": true,
+    "summary_activation": null,
+    "summary_first_dropout": 0.1,
+    "summary_proj_to_labels": true,
+    "summary_type": "cls_index",
+    "summary_use_proj": true,
+    "task_specific_params": {
+      "text-generation": {
+        "do_sample": true,
+        "max_length": 50
+      }
+    },
+    "torch_dtype": "float32",
+    "transformers_version": "4.35.2",
+    "use_cache": true,
+    "vocab_size": 50260
+  }

checkpoint/generation_config.json

@@ -0,0 +1,6 @@
+{
+    "_from_model_config": true,
+    "bos_token_id": 50256,
+    "eos_token_id": 50256,
+    "transformers_version": "4.35.2"
+  }

checkpoint/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1073341caebe0161048436239014844c08e8e90c876a716243de40d748bc1a8
+size 497783424

checkpoint/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae943840e0a03905d0b8308e3f6decacf098823f9d5b3760d993e74d2858dc96
+size 995660293

checkpoint/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54e75e628b60bbf47c975189df6cadd2954bdb738cdab1ed51bee943916d8cde
+size 14575

checkpoint/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e5e77b611783d805064e78066c7980b99b7e814294e237573ce5feea04f3061
+size 627

checkpoint/trainer_state.json

@@ -0,0 +1,37 @@
+{
+    "best_metric": null,
+    "best_model_checkpoint": null,
+    "epoch": 0.999722914934885,
+    "eval_steps": 1000,
+    "global_step": 1804,
+    "is_hyper_param_search": false,
+    "is_local_process_zero": true,
+    "is_world_process_zero": true,
+    "log_history": [
+      {
+        "epoch": 0.28,
+        "learning_rate": 3.6141906873614186e-05,
+        "loss": 4.4207,
+        "step": 500
+      },
+      {
+        "epoch": 0.55,
+        "learning_rate": 2.2283813747228384e-05,
+        "loss": 3.4761,
+        "step": 1000
+      },
+      {
+        "epoch": 0.83,
+        "learning_rate": 8.425720620842573e-06,
+        "loss": 3.4196,
+        "step": 1500
+      }
+    ],
+    "logging_steps": 500,
+    "max_steps": 1804,
+    "num_train_epochs": 1,
+    "save_steps": 500,
+    "total_flos": 2368420383744000.0,
+    "trial_name": null,
+    "trial_params": null
+  }

checkpoint/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be9eb2e69f263713b25dc91833537b35ea0d88d98ff9893d9d11d60a5a703bb9
+size 4155

dockerfile

@@ -0,0 +1,14 @@
+# read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM python:3.9
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+COPY . .
+
+CMD ["uvicorn", "fastapi:app", "--host", "0.0.0.0", "--port", "8081"]

fastapi.py

@@ -0,0 +1,48 @@
+from fastapi import FastAPI
+
+from transformers import LineByLineTextDataset
+from transformers import DataCollatorForLanguageModeling
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
+from transformers import Trainer, TrainingArguments
+from fastapi.middleware.cors import CORSMiddleware
+app = FastAPI()
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+def load_model(model_path):
+    model = GPT2LMHeadModel.from_pretrained(model_path)
+    return model
+
+
+def load_tokenizer(tokenizer_path):
+    tokenizer = GPT2Tokenizer.from_pretrained(tokenizer_path)
+    return tokenizer
+
+model_path = r'../checkpoint/'
+model = load_model(model_path)
+tokenizer = load_tokenizer(model_path)
+
+
+def generate_text(sequence, max_new_tokens):
+    ids = tokenizer.encode(f'{sequence}', return_tensors='pt')
+    input_length = ids.size(1)
+    max_length = input_length + max_new_tokens
+    final_outputs = model.generate(
+        ids,
+        do_sample=True,
+        max_length=max_length,
+        pad_token_id=model.config.eos_token_id
+    )
+    return tokenizer.decode(final_outputs[0], skip_special_tokens=True)
+
+
+
+@app.get("/subject/{prompt}")
+async def root(prompt: str):
+    print(prompt)
+    return {"subject": generate_text("Email : " + prompt + " Subject : ", 7).split('Subject : ')[1]}