deeksonparlma
commited on
Commit
·
1a88edc
1
Parent(s):
15e0607
migrate model to pickle
Browse files- .~lock.mental_health_bot.xlsx# +0 -1
- model.ipynb +32 -32
- model.pkl +3 -0
- vectorizer.pkl +3 -0
.~lock.mental_health_bot.xlsx#
DELETED
@@ -1 +0,0 @@
|
|
1 |
-
,dickson,dickson,20.02.2023 22:13,file:///home/dickson/.config/libreoffice/4;
|
|
|
|
model.ipynb
CHANGED
@@ -2,10 +2,19 @@
|
|
2 |
"cells": [
|
3 |
{
|
4 |
"cell_type": "code",
|
5 |
-
"execution_count":
|
6 |
"id": "ace57031",
|
7 |
"metadata": {},
|
8 |
"outputs": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
{
|
10 |
"data": {
|
11 |
"text/html": [
|
@@ -83,7 +92,7 @@
|
|
83 |
"4 When healing from mental illness, early identi... "
|
84 |
]
|
85 |
},
|
86 |
-
"execution_count":
|
87 |
"metadata": {},
|
88 |
"output_type": "execute_result"
|
89 |
}
|
@@ -96,8 +105,11 @@
|
|
96 |
"import pandas as pd\n",
|
97 |
"import numpy as np\n",
|
98 |
"import torch\n",
|
99 |
-
"
|
100 |
-
"
|
|
|
|
|
|
|
101 |
"# notebook_login()\n",
|
102 |
"# Step 1: Collect and preprocess data\n",
|
103 |
"# Get all the questions from Questions column and responses from Questions column in the dataset data.csv\n",
|
@@ -129,7 +141,7 @@
|
|
129 |
},
|
130 |
{
|
131 |
"cell_type": "code",
|
132 |
-
"execution_count":
|
133 |
"id": "8f51e39d",
|
134 |
"metadata": {},
|
135 |
"outputs": [
|
@@ -150,7 +162,7 @@
|
|
150 |
},
|
151 |
{
|
152 |
"cell_type": "code",
|
153 |
-
"execution_count":
|
154 |
"id": "1d697a39",
|
155 |
"metadata": {},
|
156 |
"outputs": [
|
@@ -178,7 +190,7 @@
|
|
178 |
},
|
179 |
{
|
180 |
"cell_type": "code",
|
181 |
-
"execution_count":
|
182 |
"id": "c5dde0e4",
|
183 |
"metadata": {},
|
184 |
"outputs": [
|
@@ -202,7 +214,10 @@
|
|
202 |
"# \"You can manage your anxiety through techniques such as deep breathing, meditation, and therapy.\",\n",
|
203 |
"# \"Treatments for bipolar disorder include medication, therapy, and lifestyle changes.\"]\n",
|
204 |
"\n",
|
205 |
-
"
|
|
|
|
|
|
|
206 |
"X = vectorizer.fit_transform(questions)\n",
|
207 |
"y = responses\n",
|
208 |
"\n",
|
@@ -210,14 +225,15 @@
|
|
210 |
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n",
|
211 |
"\n",
|
212 |
"# Step 3: Choose a machine learning algorithm\n",
|
213 |
-
"model = LogisticRegression()\n",
|
|
|
214 |
"\n",
|
215 |
"# Step 4: Train the model\n",
|
216 |
"model.fit(X_train, y_train)\n",
|
217 |
"\n",
|
218 |
"# model.push_to_hub(\"tabibu-ai/mental-health-chatbot\")\n",
|
219 |
-
"pt_model = DistilBertForSequenceClassification.from_pretrained(\"model.ipynb\", from_tf=True)\n",
|
220 |
-
"pt_model.save_pretrained(\"model.
|
221 |
"# load model from hub\n",
|
222 |
"\n",
|
223 |
"# Step 5: Evaluate the model\n",
|
@@ -231,36 +247,20 @@
|
|
231 |
},
|
232 |
{
|
233 |
"cell_type": "code",
|
234 |
-
"execution_count":
|
235 |
"id": "14406312",
|
236 |
"metadata": {},
|
237 |
-
"outputs": [
|
238 |
-
{
|
239 |
-
"name": "stdout",
|
240 |
-
"output_type": "stream",
|
241 |
-
"text": [
|
242 |
-
"Ask me anything : I feel sad\n"
|
243 |
-
]
|
244 |
-
}
|
245 |
-
],
|
246 |
"source": [
|
247 |
-
"new_question =
|
248 |
]
|
249 |
},
|
250 |
{
|
251 |
"cell_type": "code",
|
252 |
-
"execution_count":
|
253 |
"id": "6b9198db",
|
254 |
"metadata": {},
|
255 |
-
"outputs": [
|
256 |
-
{
|
257 |
-
"name": "stdout",
|
258 |
-
"output_type": "stream",
|
259 |
-
"text": [
|
260 |
-
"Prediction: ['\"It is estimated that mental illness affects 1 in 5 adults in America']\n"
|
261 |
-
]
|
262 |
-
}
|
263 |
-
],
|
264 |
"source": [
|
265 |
"new_question_vector = vectorizer.transform([new_question])\n",
|
266 |
"prediction = model.predict(new_question_vector)\n",
|
|
|
2 |
"cells": [
|
3 |
{
|
4 |
"cell_type": "code",
|
5 |
+
"execution_count": 1,
|
6 |
"id": "ace57031",
|
7 |
"metadata": {},
|
8 |
"outputs": [
|
9 |
+
{
|
10 |
+
"name": "stderr",
|
11 |
+
"output_type": "stream",
|
12 |
+
"text": [
|
13 |
+
"2023-02-21 17:41:49.330107: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory\n",
|
14 |
+
"2023-02-21 17:41:49.330992: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory\n",
|
15 |
+
"2023-02-21 17:41:49.331010: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n"
|
16 |
+
]
|
17 |
+
},
|
18 |
{
|
19 |
"data": {
|
20 |
"text/html": [
|
|
|
92 |
"4 When healing from mental illness, early identi... "
|
93 |
]
|
94 |
},
|
95 |
+
"execution_count": 1,
|
96 |
"metadata": {},
|
97 |
"output_type": "execute_result"
|
98 |
}
|
|
|
105 |
"import pandas as pd\n",
|
106 |
"import numpy as np\n",
|
107 |
"import torch\n",
|
108 |
+
"import pickle\n",
|
109 |
+
"\n",
|
110 |
+
"# from transformers import AutoTokenizer, AutoModelForSequenceClassification,BertTokenizer, TFBertForSequenceClassification\n",
|
111 |
+
"# from huggingface_hub import notebook_login\n",
|
112 |
+
"import tensorflow as tf\n",
|
113 |
"# notebook_login()\n",
|
114 |
"# Step 1: Collect and preprocess data\n",
|
115 |
"# Get all the questions from Questions column and responses from Questions column in the dataset data.csv\n",
|
|
|
141 |
},
|
142 |
{
|
143 |
"cell_type": "code",
|
144 |
+
"execution_count": 8,
|
145 |
"id": "8f51e39d",
|
146 |
"metadata": {},
|
147 |
"outputs": [
|
|
|
162 |
},
|
163 |
{
|
164 |
"cell_type": "code",
|
165 |
+
"execution_count": 9,
|
166 |
"id": "1d697a39",
|
167 |
"metadata": {},
|
168 |
"outputs": [
|
|
|
190 |
},
|
191 |
{
|
192 |
"cell_type": "code",
|
193 |
+
"execution_count": 11,
|
194 |
"id": "c5dde0e4",
|
195 |
"metadata": {},
|
196 |
"outputs": [
|
|
|
214 |
"# \"You can manage your anxiety through techniques such as deep breathing, meditation, and therapy.\",\n",
|
215 |
"# \"Treatments for bipolar disorder include medication, therapy, and lifestyle changes.\"]\n",
|
216 |
"\n",
|
217 |
+
"\n",
|
218 |
+
"\n",
|
219 |
+
"# vectorizer = TfidfVectorizer()\n",
|
220 |
+
"vectorizer = pickle.load(open(\"vectorizer.pkl\", \"rb\"))\n",
|
221 |
"X = vectorizer.fit_transform(questions)\n",
|
222 |
"y = responses\n",
|
223 |
"\n",
|
|
|
225 |
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n",
|
226 |
"\n",
|
227 |
"# Step 3: Choose a machine learning algorithm\n",
|
228 |
+
"# model = LogisticRegression()\n",
|
229 |
+
"model = pickle.load(open(\"model.pkl\", \"rb\"))\n",
|
230 |
"\n",
|
231 |
"# Step 4: Train the model\n",
|
232 |
"model.fit(X_train, y_train)\n",
|
233 |
"\n",
|
234 |
"# model.push_to_hub(\"tabibu-ai/mental-health-chatbot\")\n",
|
235 |
+
"# pt_model = DistilBertForSequenceClassification.from_pretrained(\"model.ipynb\", from_tf=True)\n",
|
236 |
+
"# pt_model.save_pretrained(\"model.pt\")\n",
|
237 |
"# load model from hub\n",
|
238 |
"\n",
|
239 |
"# Step 5: Evaluate the model\n",
|
|
|
247 |
},
|
248 |
{
|
249 |
"cell_type": "code",
|
250 |
+
"execution_count": 12,
|
251 |
"id": "14406312",
|
252 |
"metadata": {},
|
253 |
+
"outputs": [],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
254 |
"source": [
|
255 |
+
"new_question = \"What are the symptoms of depression?\"\n"
|
256 |
]
|
257 |
},
|
258 |
{
|
259 |
"cell_type": "code",
|
260 |
+
"execution_count": null,
|
261 |
"id": "6b9198db",
|
262 |
"metadata": {},
|
263 |
+
"outputs": [],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
264 |
"source": [
|
265 |
"new_question_vector = vectorizer.transform([new_question])\n",
|
266 |
"prediction = model.predict(new_question_vector)\n",
|
model.pkl
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9b1f64b718871770f1106973cc217f454b421b8f8a78daa12c743b43878448cf
|
3 |
+
size 2118910
|
vectorizer.pkl
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa0a144d5a7bac919c542bc480830f32bc427b541482de601ad9c87dd095c976
|
3 |
+
size 26406
|