Tru¶
Bases: SingletonPerName
Tru is the main class that provides an entry points to trulens-eval. Tru lets you:
- Log chain prompts and outputs
- Log chain Metadata
- Run and log feedback functions
- Run streamlit dashboard to view experiment results
All data is logged to the current working directory to default.sqlite.
Source code in trulens_eval/trulens_eval/tru.py
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 |
|
Chain(*args, **kwargs)
¶
Create a TruChain with database managed by self.
Source code in trulens_eval/trulens_eval/tru.py
40 41 42 43 44 45 46 47 |
|
__init__()
¶
TruLens instrumentation, logging, and feedback functions for chains. Creates a local database 'default.sqlite' in current working directory.
Source code in trulens_eval/trulens_eval/tru.py
49 50 51 52 53 54 55 56 57 58 59 |
|
add_chain(chain_json, chain_id=None)
¶
Add a chain to the database.
Source code in trulens_eval/trulens_eval/tru.py
179 180 181 182 183 184 185 186 |
|
add_feedback(result_json)
¶
Add a single feedback result to the database.
Source code in trulens_eval/trulens_eval/tru.py
188 189 190 191 192 193 194 195 196 197 198 199 |
|
add_feedbacks(result_jsons)
¶
Add multiple feedback results to the database.
Source code in trulens_eval/trulens_eval/tru.py
201 202 203 204 205 206 207 |
|
add_record(prompt, response, record_json, tags='', ts=None, total_tokens=None, total_cost=None)
¶
Add a record to the database.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
prompt |
str
|
Chain input or "prompt". |
required |
response |
str
|
Chain output or "response". |
required |
record_json |
JSON
|
Record as produced by |
required |
tags |
str
|
Additional metadata to include with the record. |
''
|
ts |
int
|
Timestamp of record creation. |
None
|
total_tokens |
int
|
The number of tokens generated in |
None
|
total_cost |
float
|
The cost of producing the response. |
None
|
Returns:
Name | Type | Description |
---|---|---|
str | Unique record identifier. |
Source code in trulens_eval/trulens_eval/tru.py
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 |
|
get_chain(chain_id)
¶
Look up a chain from the database.
Source code in trulens_eval/trulens_eval/tru.py
209 210 211 212 213 214 |
|
get_records_and_feedback(chain_ids)
¶
Get records, their feeback results, and feedback names from the database.
Source code in trulens_eval/trulens_eval/tru.py
216 217 218 219 220 221 222 223 |
|
reset_database()
¶
Reset the database. Clears all tables.
Source code in trulens_eval/trulens_eval/tru.py
61 62 63 64 65 66 |
|
run_dashboard(_dev=False)
¶
Runs a streamlit dashboard to view logged results and chains
Raises:
Type | Description |
---|---|
ValueError
|
Dashboard is already running. |
Returns:
Name | Type | Description |
---|---|---|
Process |
Process
|
Process containing streamlit dashboard. |
Source code in trulens_eval/trulens_eval/tru.py
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 |
|
run_feedback_functions(record_json, feedback_functions, chain_json=None)
¶
Run a collection of feedback functions and report their result.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
record_json |
JSON
|
The record on which to evaluate the feedback |
required |
chain_json |
JSON
|
The chain that produced the given record. |
None
|
feedback_functions |
Sequence[Feedback]
|
A collection of feedback |
required |
Returns nothing.
Source code in trulens_eval/trulens_eval/tru.py
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 |
|
start_evaluator(fork=False)
¶
Start a deferred feedback function evaluation thread.
Source code in trulens_eval/trulens_eval/tru.py
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 |
|
stop_dashboard()
¶
Stop existing dashboard if running.
Raises:
Type | Description |
---|---|
ValueError
|
Dashboard is already running. |
Source code in trulens_eval/trulens_eval/tru.py
282 283 284 285 286 287 288 289 290 291 292 |
|
stop_evaluator()
¶
Stop the deferred feedback evaluation thread.
Source code in trulens_eval/trulens_eval/tru.py
264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 |
|