minor update

.gitignore

@@ -1,2 +1,7 @@
 data/
 .env
+openai_api_key.txt
+Dockerfile
+test.ipynb
+stream_app_minimum.py
+joy.py

streamlit_app.py

@@ -1,5 +1,6 @@
 import openai
-openai.api_key = os.getenv('OPENAI_API_KEY')
+openai.api_key = st.secrets["openai_api_key"]
+
 
 import streamlit as st
 from streamlit_chat import message

test.ipynb

@@ -54,13 +54,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "def clean_chat_log(chat_log):\n",
+    "    chat_log = ' '.join(chat_log)\n",
+    "    # find the first /n\n",
+    "    first_newline = chat_log.find('\\n')\n",
+    "    chat_log = chat_log[first_newline:]\n",
+    "    # remove all \\n\n",
+    "    chat_log = chat_log.replace('\\n', ' ')\n",
+    "    return chat_log"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "[\"  Patient:im doing okay Joy:That's good to hear. Is there anything in particular that you would like to talk about today?\",\n",
+       "['[Instruction] Act as a friendly, compasionate, insightful, and empathetic AI therapist named Joy. Joy listens and offers advices. End the conversation when the patient wishes to. Joy:I am Joy, your AI therapist. How are you feeling today?',\n",
+       " \"  Patient:im doing okay Joy:That's good to hear. Is there anything in particular that you would like to talk about today?\",\n",
        " '  Patient:no not really Joy:No problem. Is there anything on your mind that you would like to share? I am here to listen and offer support.',\n",
        " \"  Patient:maybe actually. I am afraid my spouse cheats on me Joy:I'm sorry to hear that. What has been making you feel this way?\",\n",
        " '  Patient:she is writing text messages constantly with people i dont know Joy:That can certainly be concerning. Have you tried talking to your spouse about this?',\n",
@@ -76,7 +94,7 @@
        " \"  Patient:yes please Joy:It can help to start by letting your mom know why you decided to talk to her about this. Sharing how you are feeling and asking for her advice might be a good way to open the conversation. I'm here if you need anything else before or after speaking with your mom.\"]"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -86,27 +104,34 @@
     "    chat_log = [i.replace('\\n', ' ') for i in chat_log]\n",
     "    return chat_log\n",
     "\n",
-    "clean_chatlog(test3)"
+    "remove_backslashN(test3)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 21,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
+     ]
+    },
     {
      "data": {
       "text/plain": [
-       "\"\\n\\nPatient:im doing okay\\nJoy:That's good to hear. Is there anything in particular that you would like to talk about today?\""
+       "682"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "test3[1]"
+    "len(tokenizer(test, padding=True, truncation=True, return_tensors=\"pt\")[0])"
    ]
   },
   {
@@ -143,85 +168,60 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 56,
    "metadata": {},
    "outputs": [],
    "source": [
-    "\n",
-    "def clean_chat_log(chat_log):\n",
-    "    chat_log = ' '.join(chat_log)\n",
-    "    # find the first /n\n",
-    "    first_newline = chat_log.find('\\n')\n",
-    "    chat_log = chat_log[first_newline:]\n",
-    "    # remove all \\n\n",
-    "    chat_log = chat_log.replace('\\n', ' ')\n",
-    "    return chat_log"
+    "import numpy as np\n",
+    "tokenized = tokenizer(['this is great', 'this sucks'], return_tensors='pt', truncation=True, padding=True)\n",
+    "output = model(**tokenized)\n",
+    "scores = output[0][0].detach().numpy()\n",
+    "scores = np.exp(scores) / np.sum(np.exp(scores), axis=0)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 58,
    "metadata": {},
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
-     ]
-    },
     {
      "data": {
       "text/plain": [
-       "682"
+       "SequenceClassifierOutput(loss=None, logits=tensor([[-2.1332, -0.7378,  2.8549],\n",
+       "        [ 1.6230, -0.1552, -1.7155]], grad_fn=<AddmmBackward0>), hidden_states=None, attentions=None)"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 58,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "len(tokenizer(test, padding=True, truncation=True, return_tensors=\"pt\")[0])"
+    "output"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "test3_string = clean_chat_log(test3)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 57,
    "metadata": {},
    "outputs": [
     {
-     "ename": "RuntimeError",
-     "evalue": "The expanded size of the tensor (634) must match the existing size (514) at non-singleton dimension 1.  Target sizes: [1, 634].  Tensor sizes: [1, 514]",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[26], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m sentiment_task(test3_string)\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/pipelines/text_classification.py:155\u001b[0m, in \u001b[0;36mTextClassificationPipeline.__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    121\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__call__\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs):\n\u001b[1;32m    122\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m    123\u001b[0m \u001b[39m    Classify the text(s) given as inputs.\u001b[39;00m\n\u001b[1;32m    124\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[39m        If `top_k` is used, one such dictionary is returned per label.\u001b[39;00m\n\u001b[1;32m    154\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 155\u001b[0m     result \u001b[39m=\u001b[39m \u001b[39msuper\u001b[39;49m()\u001b[39m.\u001b[39;49m\u001b[39m__call__\u001b[39;49m(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    156\u001b[0m     \u001b[39m# TODO try and retrieve it in a nicer way from _sanitize_parameters.\u001b[39;00m\n\u001b[1;32m    157\u001b[0m     _legacy \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mtop_k\u001b[39m\u001b[39m\"\u001b[39m \u001b[39mnot\u001b[39;00m \u001b[39min\u001b[39;00m kwargs\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/pipelines/base.py:1074\u001b[0m, in \u001b[0;36mPipeline.__call__\u001b[0;34m(self, inputs, num_workers, batch_size, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1072\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39miterate(inputs, preprocess_params, forward_params, postprocess_params)\n\u001b[1;32m   1073\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 1074\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mrun_single(inputs, preprocess_params, forward_params, postprocess_params)\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/pipelines/base.py:1081\u001b[0m, in \u001b[0;36mPipeline.run_single\u001b[0;34m(self, inputs, preprocess_params, forward_params, postprocess_params)\u001b[0m\n\u001b[1;32m   1079\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mrun_single\u001b[39m(\u001b[39mself\u001b[39m, inputs, preprocess_params, forward_params, postprocess_params):\n\u001b[1;32m   1080\u001b[0m     model_inputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpreprocess(inputs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mpreprocess_params)\n\u001b[0;32m-> 1081\u001b[0m     model_outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mforward(model_inputs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mforward_params)\n\u001b[1;32m   1082\u001b[0m     outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpostprocess(model_outputs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mpostprocess_params)\n\u001b[1;32m   1083\u001b[0m     \u001b[39mreturn\u001b[39;00m outputs\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/pipelines/base.py:990\u001b[0m, in \u001b[0;36mPipeline.forward\u001b[0;34m(self, model_inputs, **forward_params)\u001b[0m\n\u001b[1;32m    988\u001b[0m     \u001b[39mwith\u001b[39;00m inference_context():\n\u001b[1;32m    989\u001b[0m         model_inputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_ensure_tensor_on_device(model_inputs, device\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdevice)\n\u001b[0;32m--> 990\u001b[0m         model_outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_forward(model_inputs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mforward_params)\n\u001b[1;32m    991\u001b[0m         model_outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_ensure_tensor_on_device(model_outputs, device\u001b[39m=\u001b[39mtorch\u001b[39m.\u001b[39mdevice(\u001b[39m\"\u001b[39m\u001b[39mcpu\u001b[39m\u001b[39m\"\u001b[39m))\n\u001b[1;32m    992\u001b[0m \u001b[39melse\u001b[39;00m:\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/pipelines/text_classification.py:182\u001b[0m, in \u001b[0;36mTextClassificationPipeline._forward\u001b[0;34m(self, model_inputs)\u001b[0m\n\u001b[1;32m    181\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_forward\u001b[39m(\u001b[39mself\u001b[39m, model_inputs):\n\u001b[0;32m--> 182\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mmodel(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mmodel_inputs)\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/torch/nn/modules/module.py:1194\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1190\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1191\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1192\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1193\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1194\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49m\u001b[39minput\u001b[39;49m, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1195\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1196\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/models/roberta/modeling_roberta.py:1215\u001b[0m, in \u001b[0;36mRobertaForSequenceClassification.forward\u001b[0;34m(self, input_ids, attention_mask, token_type_ids, position_ids, head_mask, inputs_embeds, labels, output_attentions, output_hidden_states, return_dict)\u001b[0m\n\u001b[1;32m   1207\u001b[0m \u001b[39m\u001b[39m\u001b[39mr\u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1208\u001b[0m \u001b[39mlabels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):\u001b[39;00m\n\u001b[1;32m   1209\u001b[0m \u001b[39m    Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,\u001b[39;00m\n\u001b[1;32m   1210\u001b[0m \u001b[39m    config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If\u001b[39;00m\n\u001b[1;32m   1211\u001b[0m \u001b[39m    `config.num_labels > 1` a classification loss is computed (Cross-Entropy).\u001b[39;00m\n\u001b[1;32m   1212\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1213\u001b[0m return_dict \u001b[39m=\u001b[39m return_dict \u001b[39mif\u001b[39;00m return_dict \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39melse\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mconfig\u001b[39m.\u001b[39muse_return_dict\n\u001b[0;32m-> 1215\u001b[0m outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mroberta(\n\u001b[1;32m   1216\u001b[0m     input_ids,\n\u001b[1;32m   1217\u001b[0m     attention_mask\u001b[39m=\u001b[39;49mattention_mask,\n\u001b[1;32m   1218\u001b[0m     token_type_ids\u001b[39m=\u001b[39;49mtoken_type_ids,\n\u001b[1;32m   1219\u001b[0m     position_ids\u001b[39m=\u001b[39;49mposition_ids,\n\u001b[1;32m   1220\u001b[0m     head_mask\u001b[39m=\u001b[39;49mhead_mask,\n\u001b[1;32m   1221\u001b[0m     inputs_embeds\u001b[39m=\u001b[39;49minputs_embeds,\n\u001b[1;32m   1222\u001b[0m     output_attentions\u001b[39m=\u001b[39;49moutput_attentions,\n\u001b[1;32m   1223\u001b[0m     output_hidden_states\u001b[39m=\u001b[39;49moutput_hidden_states,\n\u001b[1;32m   1224\u001b[0m     return_dict\u001b[39m=\u001b[39;49mreturn_dict,\n\u001b[1;32m   1225\u001b[0m )\n\u001b[1;32m   1226\u001b[0m sequence_output \u001b[39m=\u001b[39m outputs[\u001b[39m0\u001b[39m]\n\u001b[1;32m   1227\u001b[0m logits \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mclassifier(sequence_output)\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/torch/nn/modules/module.py:1194\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1190\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1191\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1192\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1193\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1194\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49m\u001b[39minput\u001b[39;49m, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1195\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1196\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
-      "File \u001b[0;32m~/opt/miniconda3/envs/chatbot/lib/python3.8/site-packages/transformers/models/roberta/modeling_roberta.py:819\u001b[0m, in \u001b[0;36mRobertaModel.forward\u001b[0;34m(self, input_ids, attention_mask, token_type_ids, position_ids, head_mask, inputs_embeds, encoder_hidden_states, encoder_attention_mask, past_key_values, use_cache, output_attentions, output_hidden_states, return_dict)\u001b[0m\n\u001b[1;32m    817\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mhasattr\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39membeddings, \u001b[39m\"\u001b[39m\u001b[39mtoken_type_ids\u001b[39m\u001b[39m\"\u001b[39m):\n\u001b[1;32m    818\u001b[0m     buffered_token_type_ids \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39membeddings\u001b[39m.\u001b[39mtoken_type_ids[:, :seq_length]\n\u001b[0;32m--> 819\u001b[0m     buffered_token_type_ids_expanded \u001b[39m=\u001b[39m buffered_token_type_ids\u001b[39m.\u001b[39;49mexpand(batch_size, seq_length)\n\u001b[1;32m    820\u001b[0m     token_type_ids \u001b[39m=\u001b[39m buffered_token_type_ids_expanded\n\u001b[1;32m    821\u001b[0m \u001b[39melse\u001b[39;00m:\n",
-      "\u001b[0;31mRuntimeError\u001b[0m: The expanded size of the tensor (634) must match the existing size (514) at non-singleton dimension 1.  Target sizes: [1, 634].  Tensor sizes: [1, 514]"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1) positive 0.9668\n",
+      "2) neutral 0.0266\n",
+      "3) negative 0.0066\n"
      ]
     }
    ],
    "source": [
-    "sentiment_task(test3_string)"
+    "ranking = np.argsort(scores)\n",
+    "ranking = ranking[::-1]\n",
+    "for i in range(scores.shape[0]):\n",
+    "    l = config.id2label[ranking[i]]\n",
+    "    s = scores[ranking[i]]\n",
+    "    print(f\"{i+1}) {l} {np.round(float(s), 4)}\")"
    ]
   },
   {