add project files

adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "openai-community/gpt2-medium",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "c_attn"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

app.py

@@ -0,0 +1,48 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
+import streamlit as st
+tokenizer = AutoTokenizer.from_pretrained("SSahas/openai_community_med_e3")
+model = AutoModelForCausalLM.from_pretrained("SSahas/openai_community_med_e3")
+
+
+def response_generator(prompt):
+    input_text = tokenizer.apply_chat_template(
+        prompt, tokenize=False, truncation=False, add_generation_prompt=True)
+    print(input_text)
+    input_ids = tokenizer(input_text, padding=True, return_tensors="pt")
+    output_ids = model.generate(input_ids=input_ids['input_ids'], generation_config=GenerationConfig(
+        max_new_tokens=20, pad_token_id=50256))
+    output = tokenizer.decode(
+        output_ids[0][input_ids['input_ids'].shape[1]:], skip_special_tokens=True)
+
+    return output
+
+
+st.title("Simple firendly chatbot")
+
+# Initialize chat history
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
+# Display chat messages from history on app rerun
+for message in st.session_state.messages:
+    with st.chat_message(message["role"]):
+        st.markdown(message["content"])
+
+# Accept user input
+if prompt := st.chat_input("What is up?"):
+    # Add user message to chat history
+    st.session_state.messages.append({"role": "user", "content": prompt})
+    # Display user message in chat message container
+    with st.chat_message("user"):
+        st.markdown(prompt)
+
+    # Display assistant response in chat message container
+    with st.chat_message("assistant"):
+        # response = st.write(response_generator(prompt))
+        # print(prompt)
+        print(st.session_state.messages)
+        response = response_generator(st.session_state.messages)
+        st.write(response)
+    # Add assistant response to chat history
+    st.session_state.messages.append(
+        {"role": "assistant", "content": response})

load_model.py

@@ -0,0 +1,44 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
+
+model_link = "SSahas/openai_community_med_e3"
+tokenizer = AutoTokenizer.from_pretrained(model_link)
+finetuned_model = AutoModelForCausalLM.from_pretrained(model_link)
+original_model  = AutoModelForCausalLM.from_pretrained("openai-community/gpt2-medium")
+
+prompt = [{'role': 'user', 'content': 'Hey man , you wanna buy some weed ?'},
+ {'role': 'assistant', 'content': 'Some what ?'},
+ {'role': 'user',
+  'content': 'Weed ! You know ? Pot , Ganja , Mary Jane some chronic !'},
+ {'role': 'assistant', 'content': 'Oh , umm , no thanks .'},
+ {'role': 'user',
+  'content': 'I also have blow if you prefer to do a few lines .'},
+ {'role': 'assistant', 'content': 'No , I am ok , really .'},
+ {'role': 'user',
+  'content': 'Come on man ! I even got dope and acid ! Try some !'},
+ {'role': 'assistant',
+  'content': 'Do you really have all of these drugs ? Where do you get them from ?'},
+ {'role': 'user',
+  'content': 'I got my connections ! Just tell me what you want and I ’ ll even give you one ounce for free .'},
+   {'role': 'assistant', 'content': 'Sounds good ! Let''s see , I want .'},
+ {'role': 'user', 'content': 'Yeah ?'}]
+
+
+prompt = [{'role': 'user', 'content': 'Hello, My name is Sahas., How are you?'},]
+
+
+input_text = tokenizer.apply_chat_template(prompt,tokenize = False, truncation=False, add_generation_prompt=True)
+    #print(input_text)
+input_ids = tokenizer(input_text,padding = True, return_tensors = "pt")
+#print(input_ids)
+#output = model.generate(input_ids=input_ids['input_ids'], generation_config=GenerationConfig(max_new_tokens=25,temperature = 0.1, eos_token_id = 50256, repetition_penalty = 1.9, do_sample= True))
+finetuned_model_output = finetuned_model.generate(input_ids=input_ids['input_ids'], generation_config=GenerationConfig(max_new_tokens=20,pad_token_id = 50256, temperature = 0.5,  do_sample= True))
+#print(output)
+original_model_output = original_model.generate(input_ids=input_ids['input_ids'], generation_config=GenerationConfig(max_new_tokens=20, temperature = 0.5,  do_sample= True))
+finetuned_model_output = tokenizer.decode(finetuned_model_output[0][input_ids['input_ids'].shape[1]:], skip_special_tokens=True)
+original_model_output = tokenizer.decode(original_model_output[0][input_ids['input_ids'].shape[1]:], skip_special_tokens=True)
+
+print("finetuned_model outptut\\n")
+print(finetuned_model_output)
+print("original_model outptut\\n")
+print(original_model_output)
+

openai_community_med_e3

@@ -0,0 +1 @@
+Subproject commit f155ca7b092859d25da29621fccd48ba0bf3dbc0